Submission on Capital Review

www.bankomb.org.nz | help@bankomb.org.nz | 0800 805 950 Freepost 218 002 | PO BOX 25 327 | Featherston St | Wellington 6146 14 May 2019 Simone Robbers, Susan Guthrie, Ian Woolford Reserve Bank of New Zealand PO Box 2498 Wellington 6140 Email: CapitalReview@rbnz.govt.nz Dear Sir/Madam Submission on Capital Review Thank you for the opportunity to comment on the proposed changes to minimum capital requirements for banks. The Banking Ombudsman Scheme is an approved dispute resolution scheme under the Financial Service Providers (Registration and Dispute Resolution) Act 2008. Our participants are registered banks and their subsidiaries and related companies, and non-bank deposit takers that meet certain criteria. Our aim is to improve the banking experience for customers and banks, as well to help resolve disputes between banks and their customers. At a broader level, we are interested in any policy developments that potentially impact New Zealand banking customers. We commend the Reserve Bank for considering the overall stability of New Zealand’s financial system and the long term resilience of banks to withstand economic downturns. This is an important conversation and the settings should be regularly revisited so New Zealanders can be assured their banking system is secure and well positioned for the future. We therefore support the policy intent behind these proposals - to mitigate the risks to banking customers in the event of economic downturns and, in the worst case scenario, bank insolvency. We do not have any specific comments to make on how capital ratios could be set to achieve that long term outcome. Nor would it be appropriate for us - as a consumer complaint scheme - to do so. We would, however, encourage the Reserve Bank to consider the following long term impacts in evaluating different options:

• How the costs of the proposed changes might be passed on to banking customers and small businesses • Whether the changes may restrict banks’ appetite for higher-risk lending and how this could reduce access to safe credit for those on lower incomes • Any consequential changes to interest rates that could drive customers towards less supervised or higher risk lending and investing options. We would also encourage careful consideration of how such changes would be introduced. What we see in our cases is that even when banks make small changes to their lending and borrowing practices, the ramifications can be significant for individuals and small businesses who have made financial decisions under a different regime. We trust these general comments are helpful and look forward to further involvement as your review progresses. Yours sincerely Nicola Sladden Banking Ombudsman

From: Ben Thirkell-White To: Capital Review Subject: Brief comments on Capital Review Paper 4 Date: Wednesday, 23 January 2019 1:53:43 PM Dear Review Team, I came across some commentary on these proposals on Newsroom and went to look at the primary documents. I am also in the process of preparing a fairly long submission for the latest round of consultation on Phase 2 of the RBNZ reform process. I write as an 'ordinary member of the public' but one who is unusually well informed on the issues (I have been researching financial crises for 20 years and I have done some substantial research on the politics of regulatory responses to the GFC in the UK). I thought the paper did an excellent job of setting out the evidentiary basis for the proposed levels of capital in as a clear a manner as possible given the technical nature of the subject matter. Particularly given some negative coverage from Newsroom, I wanted to be very clear that I support the broad parameters of what is being proposed. It seems to me to strike a sensible balance between prudence and banking efficiency. In terms of risk appetite, I might be willing to contemplate a banking crisis slightly more often than once every 200 years (depending a little on how serious it was and how it was dealt with). However, I think it is appropriate to be somewhat conservative for the following reasons: (a) The costs of crisis can be extremely high and crises can force governments into bailouts that are sensible from a utilitarian point of view at the time but also distributionally highly undesirable (b) There is considerable uncertainty in modelling, especially in the NZ context where we have little historical data to work on (c) As the IMF pointed out in the 2016 FSAP, there are particular reasons for concern because of market concentration amongst 4 banks with similar business models, which could create unexpected outcomes not well captured in stress tests (d) The Reserve Bank's 'non-intrusive' supervision regime clearly increases risk of poor practice / mis-reporting. It seems reasonable that banks pay some regulatory price for their unusual autonomy in that respect. (In other words, the 200 years is notional in any case and, because of unknown unknowns, probably understates true likelihood of crisis) I'm not qualified to comment on the detail of the modelling but 16% sounds about right to me. Lord Turner and John Vickers are both on record as saying that they thought the 'right'

number for the UK was in the region of 17-20% but that they 'couldn't get it', though the figure for ring-fenced systemic banks is not wildly different to that. For those concerned that 16% is high, the proposed structure of a fairly large 'buffer' before a bank loses its license should give some comfort. Banks shouldn't need to hold as much above 16% as they might hold above '8%'. The increased scrutiny approach seems to me to align well with the incentives one would like to give bank management as capital began to appear low. It provides a transparent warning that will push directors to raise new capital. I'm entirely happy with a levy for large significant banks. One of the reasons for risk regulation is the social externalities of bank failures. The risk of that is greater for systemically significant banks. More unpalatably, the reality is that they also benefit from a greater implicit (and ambiguous etc.) government guarantee (whether one wishes to acknowledge it or not) and that justifies a slightly tighter regulatory stance. Whether the figure is correct is a bit harder to say - I think it's lower than that applied by US and UK authorities (?) but don't feel hugely qualified to debate precise quantification. Similarly, counter-cyclical capital buffers seem like a good support to the RBNZ's emerging macroprudential role, particularly in an environment in which we're all more aware of the zero lower bound problem. I don't think there's anything rocket science in all that but I thought it might be helpful to get an informed view from someone that is fully independent of the NZ financial sector. Best wishes Ben Thirkell-White Dr Ben Thirkell-White Associate Professor Political Science and International Relations Victoria University of Wellington Te Whare Wananga O Te Upoko O Te Ika A Maui PO Box 600 Wellington 6140 New Zealand OIA s9(2)(a)

Benno & Ann Schmitt Dear Mr Orr We 100% agree with your proposals to increase the amount of capital banks must have, to ensure they can withstand financial and economic shocks. Regards Benno & Ann Schmitt OIA s9(2)(a)

Benton Macdonald Please err on the side of more capital. I understand that when a bank extends a loan that money is substantially created. Nationally created money properly belongs to the people of that nation. Currently created money substantially goes to banks inflating real estate. Please change the rules so that created money benefits the people of the nation. You could for example make a rule that 25% of created money must be invested in education. OIA s9(2)(a)

From: Bill Foster To: Capital Review Subject: Bank Capital Review Date: Monday, 4 February 2019 1:24:43 PM Attachments: Bank Capital Review Submission.pdf Please find attached my submission on this subject. I am happy for the submission to be published. Best Regards

---

William Foster Capital Market and Governance Specialist OIA s9(2)(a)

CONVENTIONAL APPROACHES TO MORE PROTECTION WILL ADD COST Conventional (orthodox) approaches to the capital adequacy question are to mitigate the risk by having an acceptable level of capital available adjusted for risk. Increasing equity capital, or additional risk mitigation (more regulation; insurance; security; or guarantees), adds costs to the process of capital allocation and reduces the efficiency of that allocation. Unconventional approaches, which aim to eliminate or reduce the risks to be protected against, are not being considered in this review, but they should be. The starting point (in most western democracies) for privately owned limited liability companies is that they can fail (and should be allowed to). Large systemic financial market infrastructures have more recently been regarded as “too-big-to￾fail” with widespread acceptance that the consequences would greatly affect the economy and voters, and the Government will regulate to prevent, or be obliged to act in the event of, major bank failure. REDUCING THE RISKS TO BE MANAGED Primary amongst political concerns is the protection of ordinary depositors in a bank failure, so this review is strongly tied to the current Treasury consultation on, inter alia, whether deposit protection is needed and if so, how it should be provided.

Since banks lend primarily to retail home buyers and farmers, then if the risk of those depositors losing any money in their transaction accounts (i.e. cheque not savings or investment accounts) in a bank failure can be reduced or eliminated, then the level of equity protection required to cover other bank liabilities against potential failure is significantly reduced. Creditors who lend money as savings or investment consciously to a financial institution (including banks), to earn a return greater than the risk-free return, generally know (and certainly should) that those savings/investments are unsecured loans to the bank and at risk of loss. Generally also, as the risk increases, the return expected by the investor is higher. Banks are incentivized to demonstrate their reputation for financial solvency and reliability in order to attract those loans for investment at lower rates. If depositors transaction funds) are fully protected, it becomes politically acceptable to reduce regulatory capital requirements and allow banks to fail and default on commercial creditors. This is likely to be healthy for bank customers a competitive financial system.

NO NEED FOR MORE CAPITAL IF DEPOSITORS FUNDS ARE PROTECTED Current stress tests (despite their unreliability and subjectivity) show that NZ banks are generally sufficiently capitalized even though depositors currently have no more protection than the OBR policy, so it is not proven that raising ratios will reduce risks. If deposits are fully protected there seems to be less or no argument for increasing bank capital requirements – which would certainly increase lending costs.

HOW BEST TO PROTECT DEPOSITORS FUNDS In relation to the Treasury review I have written to recommend that depositors be protected by making their depositors’ transaction1 account funds (typically cheque accounts) sovereign monies owned by the depositor (technically by making them sub-accounts of each bank’s ESAS account at the RB). A copy of that submission and its attached paper (which gives some background references) are appended (Appendices 1 and 2). In that paper I submit that the case for additional protection of depositors by NZ banks (and by implication the need for higher bank capital ratios) has not been made, but that if it is required this method (however unorthodox it may be viewed as by traditional bankers) is the best way forward. For the purposes of bank capital, the main consequence of such a move is to remove some $60B of bank liabilities for which capital protection is currently required. The resulting increase in capital ratios would be expected to allow a reduction in bank equity levels (to the benefit of shareholders and customers) even if a higher ratio of equity to total assets (greater than 4.5%) is required for subjective or political reasons. The paper provides a discussion of how it could work and comments on the alternatives being proposed. Some may observe that making depositors’ transaction funds sovereign money owned by the depositor is much the same as providing a Government guarantee - both provide support for, and a benefit to, banks that fully protects depositors’ transaction account funds. However, removing those funds from bank’ balance sheets has the benefits of avoiding creating a contingent liability for the Government and eliminates the moral hazard problem which remains if banks are given guarantees or insurance cover for depositors’ funds, but are still responsible for their application.

1 These might be called payment or settlement accounts to describe purpose, since transfers between different savings and investment accounts within the one bank may still take place in that bank’s private monies.

As should be well understood by now that banks create private money by lending (and destroy it when loans are repaid). They do not need any depositors’ money to lend, and they cover any risk of loan default by borrowers from capital provided by shareholders equity, various quasi-equity instruments and other borrowings. Also the level of sovereign funds required for daily interbank settlements is relatively low (circa $4B). Currently banks manage their net levels required for ESAS settlement by using their own sovereign funds or by borrowing and lending amongst other banks and the RB. This can be extended to include borrowing from (and lending to) depositors’ so banks can manage their net requirement for changes in levels of sovereign funds for transaction accounts. This option for addressing bank/depositors’ risks and capital requirements is not a new concept but needs to be seriously considered in any reevaluation of bank capital and depositors’ risks that will otherwise lead to raising the costs of banks and hence their lending rates or to introducing additional and more costly intermediation or regulation. SUMMARY Making depositors’ transaction account funds sovereign funds owned by them in the manner proposed: • is a natural extension of the concept behind the current ESAS system; • capitalises on and enhances New Zealand’s lead in efficient electronic clearing and transaction systems; • has a one-off effect of reducing bank balance sheets by $60B while keeping bank reserves unchanged, thereby benefitting bank shareholders and customers alike. • Depositors get protected transaction funds; • Government gets a political pat on the back; • regulation of banks can be more focussed (whether by RB or FMA); • banks are more focussed with more separate transaction and lending businesses; • bank activities are more transparent; • banks can be allowed to fail (improving competition); • lending rates are more competitive; • the economy benefits; and everybody wins! I commend it for your earnest consideration and would be happy to provide any further clarification or respond to any further questions. Kind Regards William Foster

The idea of making depositors the legal owners of their own “sovereign” or “real” money is not a new one, as you may see from the attached note on “Protecting Depositors’ Funds”, but discussion and arguments have continued over the different ways, and the extent to which such ideas may be implemented in modern banking systems. New Zealand now has one of the most automated money transfer systems and high levels of electronic banking and money transfers in the world. Interbank transfers already take place in Real (Reserve Bank account) money. This makes the idea of making depositors electronic deposits and transfers in their own “real” money more practical than in other countries. Notes and coin in circulation are only some $800m, out of around $60B of deposits in transaction accounts. Daily real money transfers (in ESAS) are around $30B of which around $4B represent transfers between banks of net customers’ transfers.2 . New Zealand thus has an ideal opportunity for the Government to set a leading example in the protection of depositors’ monies and the provision of secure electronic transaction services. This option is superior to the alternatives of deposit insurance or a government guarantee and I urge you to include it in your considerations. Is protection of depositors needed? The introduction of Basle III capital requirements and more extensive bank monitoring has substantially reduced the (already small) risk of Registered Bank failure in New Zealand and it is not clear that any further explicit protection for banks’ depositors is currently required. The risk of losing unsecured funds however still exists for non-bank deposit taking institutions and other financial institutions taking short term deposits of public monies. Since the Reserve Bank also supervises non-bank Financial Institutions it is not clear that there is a general understanding and acceptance that monies deposited with both bank and non-bank institutions are unsecured and “at risk”. Also, there is a general concern about the efficacy of regulation of banks by the regulators (from the Australian Banking Commission enquiry) and an implicit understanding amongst the general public (reinforced by the South Canterbury Finance and previous bank bailouts in New Zealand) that the government provides an implicit guarantee of bank and non-bank depositors funds - and maybe finance companies). These lingering concerns will be resolved once and for all if depositors’ funds intended for transaction purposes (i.e. cheque account deposits, not savings or investment accounts funds) are treated as sovereign money owned by the depositor.

2 Broad numbers from Reserve Bank stats and ESAS annual report.

Transaction Services and Banking Services increasingly distinct. Ever since depositors’ money has been declared to legally belong to the bank, banks have enjoyed the availability of money they pay no interest on, to invest or support their leveraged money creation activities from which they (and their investors) profit. However, as transaction processing systems have become more efficient (and ever cheaper) banks generally have been able to develop the provision of more extensive payment and transaction services for their customers’ convenience and benefit. The Reserve Bank has also developed modern and efficient systems for processing sovereign money accounts held by financial institutions and for transactions between them. Management of the money transfers for most bank customers is cleared in the Settlement Before Interchange (SBI) system before any net ESAS transfers. SBI is managed by a separate company (Payments NZ Ltd). Electronic payment systems are increasingly ubiquitous and payment services is now a line of bank business activity that is increasingly separate from bank lending and investment activities. How it would work. The simplest implementation for making depositors transaction monies into sovereign (otherwise called “Real”) monies therefore would be to declare depositors “cheque” accounts held at banks as sub-accounts of their ESAS (Exchange Settlement Account System) accounts for each bank and owned by the depositors. The ESAS uses sovereign (real) money held in accounts at the Reserve Bank and Interbank transfers are already all in Real Money. Funds in depositors’ transaction accounts would then be sovereign monies owned by the depositor and thus fully protected in any bank failure. The change would not be obvious to most depositors and no change is required to the current services and systems for other savings accounts and credit accounts, or for credit transaction processing. These activities could continue to use bank owned money and credit created by the banks and card companies and carry associated investment and credit risks that lenders (savings depositors and credit card companies e.g. Mastercard and Visa) are expected to understand, accept, and manage. The primary lending and private money creation activities of banks would be unaffected.

Printed cheques also would continue to be used with acceptance necessarily subject to current checks that the funds are fully cleared for transfer and transferred by the transferor bank. Notes and coin (already “sovereign money”) deposited would be immediately credited to the customers transaction (cheque) account. The requirement could be implemented by legislation or Reserve Bank Regulation. For the transition, the Reserve Bank would effectively “create” sovereign money in its accounts equivalent to the amount in depositors’ accounts (increasing its balance sheet accordingly), and exactly the same amount of private money would be removed by the banks eliminating depositors’ transaction account assets and liabilities from their balance sheets - so the money supply would be unaffected. banks would continue to hold and account for the assets of depositors in their transaction accounts, but these would be depositors’ assets, not bank assets. [Banks could not be given the power to create Sovereign money, since this would usurp the Reserve Bank’s role and effectively allow them to issue themselves sovereign funds as they wanted.] Bank Reserves would remain unchanged and with banks’ current equity levels around $40B, the balance sheet reduction would increase their capital ratios and be positive for shareholders, while depositors gain the benefit of having fully protected (sovereign) funds in their transaction accounts. Funds held by banks for depositors in their transaction accounts surplus to daily requirements for transfer to other banks would earn overnight interest at the Reserve Bank at the prevailing risk-free rate3 . This could be either paid to depositors or used to offset and subsidise the costs of money transfers. After the transition, assuming banks retain their private money creation role, transfers between (bank owned money) savings or investment accounts and (depositors owned money) [cheque] transaction accounts would require banks to borrow (and repay) Real money for transfer to (and from) depositors transaction accounts (as they do now between banks at a net level). Banks making a transfer from an investment to a transaction account would need to obtain the corresponding real funds, either from their own ESAS funds or by borrowing them e.g. from the Reserve Bank (as now). Conversely a transfer from a transaction account to a savings or investment account would be treated just as a deposit of notes and coin is now.

3 This is currently OCR - 0.5% but an increase of some $50B plus in overnight funds might lead the Reserve Bank to lower the risk-free rate (even to zero) to avoid any significant interest payment obligation and to encourage depositors to lend their sovereign funds to the Bank for its other settlements.

Banks would be free to determine whether transaction services would need to be charged for or if their customers could be paid a nominal amount for risk free deposits not used for transfers4 . If depositors chose to hold significant amounts of funds long term in transaction accounts, Banks would be free to incentivise transfers to (bank owned money) savings or investment accounts through interest differences or charges (i.e. borrowing from depositors as they do now). Such transfers would allow banks to increase their own real money deposits at the Reserve Bank or to repay their borrowings of Reserve Bank funds. Alternatives CUSTOMER ACCOUNTS AT THE RESERVE BANK (OR AS A BANK OPTION) Giving depositors direct access to Sovereign money accounts at the Reserve Bank, rather than indirectly through registered banks, might seem a logical extension of the idea of “real money” accounts (on the principle that such access should not be privileged to approved institutions only), however it would face the practical problem of significantly increasing the transaction volumes through ESAS and an unnecessarily complication of the money transfer system (i.e. having to distinguish and deal with individual transfers between real and private types of money and so removing the benefits of transferring only fungible monies in the system). It also would be a significant dis-intermediation of bank’s traditional customer servicing role that would be strongly opposed by the banks and would give the Reserve Bank a set of additional customer level relationship management responsibilities and costs that it is not set up to manage and provide for as a regulator. Front-line banks are regulated by the Reserve Bank and established providers of customer services and should continue in that role. Making real money accounts at the Reserve Bank (or at their bank) optional would also create depositor confusion over the status of the money in their accounts and not significantly reduce the expectation of an implied government guarantee over private money accounts.

4 Customers might agree to lend their money to the Bank for other settlements as required at a nominal rate.

TRUST ACCOUNTS Banks could be required to hold depositor’s money in trust for the customer, rather than treat it as an unsecured loan to the bank. This would impose a range of trust obligations on banks which would be costly and would prevent them from using depositors’ funds for investment or loan, yet they would still be included in the balance sheets of the bank. This alternative would raise costs for banks and involve more changes in bank customer relationship practices. SECURING DEPOSITORS BALANCES Banks could be required to make depositors funds secured deposits. This would require banks to raise more capital to protect these funds and thus also be more costly. GOVERNMENT GUARANTEE A Government Guarantee of depositors’ funds creates the Moral Hazard risk for banks that they can use these unsecured funds without any risk to depositors of loss of the monies. That is a government protection of banks, rather than customers and may encourage banks to undertake more risky lending and investment, thus increasing the potential liability to the Crown and taxpayers. it also creates an unfunded liability of the Government which is avoided under the proposal for making depositors funds Real money. However, as there is no direct or immediate increased cost, this is the second-best option for depositor protection. DEPOSIT INSURANCE Deposit insurance to protect depositors’ funds adds an unnecessary layer of intermediation and cost (for either banks, their customers or taxpayers), and additional time and cost in the resolution of a bank failure. It is the second most expensive deposit protection alternative (after requiring banks to fully secure deposits) and therefore less preferred as a protection mechanism. However, it will be favoured and advocated by Insurance companies and banks as it adds a regulated layer of activity and cost to the payment infrastructure which benefits financial institutions by creating new activities whose costs can be recovered through fees or charges without argument. This is one reason why it has been preferred overseas, thus allowing the argument of overseas practice to be used to support its adoption here. Application to Non-Bank financial Institutions The arrangement of giving depositors access to Reserve Bank ESAS “Real Money” accounts through banks’ sub-accounts could readily be extended to other financial institutions (non-bank deposit takers) and their depositors to ensure protection of their monies deposited i.e. not lent for investment.

Conclusion This proposal is not only the least disruptive and most cost-effective solution to the deposit protection discussion, and a natural development of the current ESAS system, but will contribute substantially to the reputation of New Zealand for the security and efficiency of our money transfer systems and allow the Reserve bank to achieve its objectives of “finality”, “irrevocability” and “certainty” in the payment system. I commend it for your earnest consideration and would be happy to provide any further clarification or respond to any further questions. Kind Regards William Foster5

5 William Foster (B.E (elect). M.Comm. MInstD, MIPENZ) is a capital market and governance specialist. He is a past Managing Director of the NZSE (predecessor of the NZX), responsible for the FASTER settlement system in the 1990’s and its associated regulation for electronic transfers. He was COO of the Abu Dhabi Securities Market and is a past Executive Director of the Mudara Institute of Directors (in the DIFC), and a Vice President of the ISEEE (a not for profit organisation of past International Exchange Executives). He has spent time in Cambodia as lead expert and project manager assisting the Securities and Exchange Commission of Cambodia (SECC) and the Government of Cambodia to develop their capital market regulatory framework and review their Financial Sector development Strategy, and more recently In Sri Lanka assisting with the demutualization of the Colombo Stock Exchange. He is a founding shareholder of Efficient Market Services Ltd, the owner of the Unlisted Securities Market (USX).

Appendix 2: Protecting depositors funds Securing cheque accounts What most people don’t realise when we “deposit” money in our cheque accounts is that we are not handing it to the bank for safekeeping: rather we are lending it to the bank and we give up ownership of the money. It is recorded as a claim against the bank and used with other money of the bank for investment, loans and credit creation6 (which carries risk). So, transactions in the payment system within banks are made with private bank money. Banks act as intermediaries, not agents, on behalf of customers in the payment system. Banks and other approved financial institutions are the ones who operate the payment system and carry out lending and credit creation. The lending and credit creation role is now taken for granted, but banks were not always intermediaries and lenders. Prior to 1811 title in depositor’s accounts belonged to the depositor.7

6 As “Keister and McAndrews (2009), staff economists at the Federal Reserve Bank of New York, write: “Suppose that Bank A gives a new loan of $20 to Firm X, which continues to hold a deposit account with Bank A. Bank A does this by crediting Firm X’s account by $20. The bank now has a new asset (the loan to Firm X) and an offsetting liability (the increase in Firm X’s deposit at the bank). Importantly, Bank A still has [unchanged] reserves in its account. In other words, the loan to Firm X does not decrease Bank A’s reserve holdings at all.” Putting this differently, the bank does not lend out reserves (money) that it already owns, rather it creates new deposit money ex nihilo.”: q.v (5). 7 Carr v Carr Nov 1811. Prior to 1811 title to the money in depositors’ accounts belonged to the depositor. Lending of depositors’ money without their consent could then have been considered fraudulent. Sir William Grant ruled that since money paid into a bank deposit had been ‘paid in generally’, and not earmarked in a sealed bag (i.e., as a ‘specific deposit’) the transaction had become a loan rather than a bailment. Following that, in Foley vs. Hill [1848], in the High Court of Chancery, Lord Cottenham stated ‘There is a fallacy in likening the dealings of a banker to the case of a deposit to which in legal effect they have no sort of resemblance, money paid into a banker’s account becomes immediately a part of his general assets and he is merely a debtor for the amount’. These two judicial decisions gave legal status to the banking practice of removing depositors’ money from their accounts and lending it to others. Since then, title to depositors’ money is transferred from the depositor to the bank at the moment that the deposit is made.

Since then private money (created by private banks) has been different from government issued or sovereign money (referred to as “Real Money8 and which only the Reserve Bank has the right to issue in New Zealand 9

• this is traditionally notes and coin, but in modern payment systems includes electronically created and issued funds in Accounts at the Reserve Bank). The system currently provides a benefit to banks, since they get depositors funds intended for payment transactions (because of the significant convenience value to customers), as well as deposits intended for savings or investment, and can use that money as they wish, including to back lending through the creation of more private money deposits on which they may then earn interest. This benefit is taken by the small group of Registered Banks that owns the privilege to privately create money. It is a privilege that, due to its enormous benefits, is often originally acquired as a result of intense rent-seeking behavior.10 Private banks thus have little interest in changing the system and have generally opposed efforts to ensure that depositors accounts are owned by depositors, or that they represent government issued money.11 The linkage to payments is important because our current electronic payment system is a modern evolution of the cheque payment system. As banks held customers’ payment monies in their cheque accounts, these were naturally the starting point for retail electronic payments.

8 Fiat money issued by the Government (Reserve Bank) - irredeemable and not to be confused with commodity backed Government issued money (e.g Gold Standard) advocated by some. 9 Under the Reserve Bank Act of 1933 (and the subsequent Reserve Bank Act of 1989), the Reserve Bank became the only authority allowed to issue New Zealand currency, and Reserve Bank Notes and coins are legal tender for settling any debt. “The Legal History of Money in New Zealand: Ken Mathews, Reserve Bank Bulletin, 2003, Vol 66 No 1”. 10 Benes and Kumhof; IMF Research paper; August 2012; "The Chicago Plan revisited" 11 Various proposals have been made - 100% reserve ratios, separating deposits from credit activities, major reform of the monetary system such as the Chicago Plan, and legislation to make deposits secured. Proposals include; in the US - The American Monetary Act, and the American Monetary NEED Act HR2990 ; and in the UK in 2008 The Earl of Caithness's Private Members Bill "The Fully Secured Current Accounts Bill".

Since each Bank creates its own private money, progressive improvements have been introduced to speed up the processing of interbank payments12, thus reducing the risk of banks’ failing to meet payment obligations to each other. Banks are part of the ESAS13 system, which provides overnight interbank settlement and allows intraday settlements for high value payments at any time during the day14. These Interbank settlements (including for the net amounts owing between banks for customers' transactions) take place in Real (Sovereign) Money (i.e. amounts in accounts held by the banks [and other approved financial institutions] at the Reserve Bank). The ability to transfer money between banks is implemented at the intermediary (Bank) level only and not at the customer transaction level. Customers do not have direct access to ESAS accounts or ownership of electronic sovereign money. The Reserve Bank charges for the privilege of providing sovereign money for transactions. ESAS accounts cannot be overdrawn. To get money in their ESAS accounts, banks can either: • deposit notes and coin lodged with them by depositors; • receive payments from other ESAS members; • repo securities (at a discount) to borrow sovereign funds; • buy funds at the prevailing rate set by the Reserve Bank (OCR +1.7%); or • borrow from other ESAS members. Conversely the Bank pays interest on surplus funds (overnight) at a rate of (OCR￾0.5%). This is the overnight rate earned on “risk free money”. Individual Bank customers have no access to their own ESAS accounts and cannot earn overnight interest on their monies at this risk-free rate. With the growing availability of credit through card companies the EFT-POS15 payment systems had to be designed to allow the electronic recording of credit card company transactions as well as banking payments and the clearing within and between banks of transactions to be settled net through the ESAS system. The modern EFT-POS system allows payments to be debited from, and credited to, cash, savings and credit accounts.

12 Since in accepting money from another bank, any bank had to be satisfied that the paying bank was solvent and could honor its obligations to redeem its monies in sovereign funds, if necessary. To resolve this mutual distrust, the Reserve Bank introduced the requirement that interbank daily net transfers be made in sovereign money, in which everyone had confidence. 13 https://www.rbnz.govt.nz/markets-and-payments/esas 14 Retail payment settlements between banks are made on a net basis (after clearing) at regular intervals. The system also provides for High Value payments of individual transactions to be made directly at any time. This is referred to as Real Time Gross Settlement (RTGS). 15 http://en.wikipedia.org/wiki/EFTPOS

Bank customers are unsecured creditors of that bank, so the weakness and risk for ordinary customers in the current system of private money payments is that if any Bank fails, unsecured creditors (including customer deposits) may be called upon to meet the bank’s liabilities. Generally, this risk raises no concerns for customers, except in the event of a failure of a bank, when the Open Bank Resolution (OBR)16 policy of the Reserve Bank now applies. The OBR policy of the Reserve Bank aims to speed up the resolution of the affairs of an insolvent financial institution but provides no full protection17 for depositors’ monies for the extent to which they may be required to meet the liabilities of the bank. Furthermore, the Reserve Bank has acknowledged that “in flight” transactions (which have been entered by customers but not yet settled when the bank is declared insolvent) are not guaranteed to be settled completely in the event of the application of the OBR policy. Unless the Reserve Bank determines that they will (at the expense of other bank creditors), the payments system does not yet have finality, certainty, and irrevocability, nor is it sure to be free from disruption - basic objectives for the payment system that the Reserve Bank has but softened a little under the current Financial Market Infrastructure (FMI) objectives by removing irrevocability and focusing more on risk management, governance, transparency and stability.18. Increasing the frequency of interbank settlements (which the Reserve Bank has encouraged) further reduces payment settlement risk, but does not remove it, nor simplify the payment system. Significant and costly mitigations (through capital requirements for banks) are currently stipulated by the Reserve Bank to minimise the risk that a Bank failure will result in deposits losing their unsecured funds. Full protection for depositors requires either: • that depositors’ funds are fully secured; • that the Government guarantees depositors; • that insurance cover is provided, or • that depositors’ funds are treated as sovereign funds owned by the depositor, as Bank funds in their ESAS accounts are now.

16 https://www.rbnz.govt.nz/regulation-and-supervision/banks/open-bank-resolution 17 The offering of a “de minimus” level of guaranteed funds for customers (to be determined in each case, but examples are as small as $4500 or $500) is not a guarantee that any significant level of depositors monies will be protected. 18 https://www.rbnz.govt.nz/-/media/ReserveBank/Files/regulation-and-supervision/financial-market￾infrastructure-oversight/oversight-of-financial-market-infrastructures-in-new-zealand.pdf?la=en

While different banks and card companies compete for customers, they all agree to use the central, standard electronic payment system, which is regulated by the Reserve Bank (which has responsibility for national payment systems). Standards for the payment system are now promulgated by Payments NZ Ltd. which also provides clearing services for banks customer transactions prior to net settlement. So, there is already a high degree of standardisation, regulation and effective monopoly in the national electronic payment system built around bank accounts. It is ubiquitous in New Zealand and there is no practical competitive alternative for the public for most payments. The Government and many organisations now mandate that people in receipt of their payments have bank accounts for receipt of monies by electronic transfer. Providing Secured or Real Money Transaction Deposit Accounts for depositors’ funds will remove any need for (and arguments for) deposit insurance or Government guarantees (which add costs for depositors or taxpayers, and introduce moral hazard for bankers). Requiring electronic transactions for individual depositors to be in real money may be achieved in a variety of ways19, but implementing a Real Money payment system does not have to mean implementing more radical monetary reforms proposed such as eliminating private credit or money creation20. However the credit system is structured, extending our Real Money payment system to cover depositors transaction balances should not impact banks’ lending and borrowing. Customer credit payments and transactions can continue as at present - the system for consumer payments already distinguishes between cheque account and credit balances and payments. Bank customers can arrange overdraft facilities to credit Real Money to their transaction accounts for payments if they wish - as they do now in private bank-owned money.

19 e.g. Other options include making cheque accounts Trust Accounts (following securities industry practice), or requiring banks to make transaction deposit accounts secured funds, or even making customers transaction accounts direct Reserve Bank accounts (though this level of disintermediation would be opposed by banks). A clear separation is needed between transaction accounts (which can only earn at best the risk free rate of interest), and savings and other investment accounts (which would be monies on loan to the bank and may be invested [at risk] to earn higher returns). Implementation could be by Legislation or Reserve Bank Regulation. 20 q.v.(5)

New Zealand has shown itself to be a leader in the development and use of electronic payment systems. Clearly the time has come, now that electronic payments are ubiquitous, and we rely fully on the system’s availability and reliability, for the Reserve Bank to implement its objectives of “finality” and “certainty” fully in the system, by requiring the monies held in (and thus transactions to and from) our cheque accounts to be made in customers own Real Money. William Foster Revised December 2018

1 The Parade of the Bankers’ New Clothes Continues: 31 Flawed Claims Debunked Anat Admati and Martin Hellwig Revised December 2015 The debate on banking regulation has been dominated by flawed and misleading claims. The title of our book The Bankers New Clothes: What’s Wrong with Banking and What to Do about It (Princeton University Press, 2013, see bankersnewclothes.com) refers to flawed claims about banking and banking regulation, and the book discusses and debunks many of them. Flawed claims are still made in the policy debate, particularly in the context of proposals that banks be funded with more equity and less debt than current or new regulations would allow. Those who make the flawed claims do so without addressing our arguments, even when commenting on the book or on our other writings. Because the financial system continues to be dangerous and distorted, however, flawed claims must not win the policy debate.1

This document provides a brief account of claims that we have come across since the book was published in February, 2013. We provide brief responses, with references to more detailed discussions in the book and elsewhere.2 Many claims are asserted without any justification. Some of these claims are simply false or based on fallacious reasoning. Other claims are misleading or irrelevant, for example confusing costs and benefits to banks or bankers with costs and benefits to society, which must be the focus of policy. Still other claims are based on implausible theories that ignore important parts of reality. We first provide a list of the flawed claims that the rest of this document takes on. References to chapter numbers refer to our book. Nothing that we heard or read changes our conclusions or our strong policy recommendations.                                                               This document is a revision of a document from October 10, 2014, which debunked 28 flawed claims (which itself was a revision of an earlier document from 2013 debunking 23 flawed claims). In addition to updates and some additional discussion and references (specifically in claims 4, and 7), we have added current Claims 6, 12, and 22. We are grateful to Peter Conti-Brown, Paul Pfleiderer and Kim Schoenholz for comments on earlier drafts. 1 Others who have written to challenge flawed claims include John Cochrane (e.g., “Capital and Language” posted at http://johnhcochrane.blogspot.com/2014/09/capital-and-language html), Mark Whitehouse (e.g., “Seven Dumb Things Bankers Say,” April 5, 2013 and “Too-Big-To-Fail Myths, Goldman Sachs Edition,” May 28, 2013, both in Bloomberg View), Bloomberg View Editors (e.g., “What’s so Radical about a Safer Financial System?” April 9, 2013) and Simon Johnson (e.g., “The Impact of Higher Capital Requirements for Banks,” April 18, 2013, Economix and “Two Views on Finance.” Project Syndicate, September 29, 2014) Paul Pfleiderer has been active in the debate with academics privately and publicly (see Pfleiderer, 2014). See also the preface of the paperback edition, available at http://press.princeton.edu/chapters/p10230.pdf 2 In some cases, we give specific references to writings where flawed claims are made, but we have not attempted to find all such references. Some of the claims have come up in various discussions of the book that we have had after its publication. Aside from the book, all our other writings are available through the book website, SSRN or the website in which we have posted writings since 2010 https://www.gsb.stanford.edu/faculty-research/excessive￾leverage

2 List of Flawed Claims Claim 1: Capital is money that banks hold or set aside as a reserve, like a rainy day fund; higher capital requirements keep money out of the economy. Claim 2: Requiring banks to hold cash reserves equal to 15% or more of their assets does not make them significantly safer, and therefore even such high capital requirement would not address the key problems in banking. Claim 3: The argument for requiring banks to have substantially more equity is only based on a theoretical result called the Modigliani-Miller theorem, which says that the funding costs of a corporation are independent of the mix of debt and equity it uses. This result does not apply in the real world because its assumptions are unrealistic. Claim 4: The key insights from corporate finance are not relevant for banks because the economics of funding for banks is entirely different from that of other companies. Claim 5: Banks are special because they create money. Claim 6: Bank deposits differ from other kinds of debt because banks themselves create deposits by lending. Claim 7: Increasing equity requirements would reduce banks' ability to take people's deposits and issue short-term claims that are liquid and can be used like money. Claim 8: Increasing equity requirements would increase the funding costs of banks because investors require higher returns when investing in equity than when investing in debt. Claim 9: Increasing equity requirements would lower the banks’ return on equity (ROE) and thus make investors unwilling to invest in banks’ stocks. Claim 10: Increasing equity requirements would constrain banks so they must reduce lending. Claim 11: Increasing equity requirements would be harmful for the economy because banks would be less willing to make loans. Claim 12: Higher equity requirements would restrict banks’ ability to provide market-making services, harm market liquidity, and prevent banks from stabilizing volatile stock markets by countering adverse price movements. Claim 13: Higher equity requirements are undesirable because they would prevent banks from taking advantage of government subsidies and thus force them to charge higher interest on loans. Claim 14: Historically, banks have never had as much as 30% equity; requiring as much equity would therefore harm the business of banking.

3 Claim 15: There is not enough equity around for banks to be funding with 30% equity. Claim 16: Because banks cannot raise equity, they will have to shrink if equity requirements are increased, and this will be bad for the economy. Claim 17: Increasing equity requirements would harm economic growth. Claim 18: Basel III is already very tough, doubling or tripling previous requirements; banks that comply with Basel III requirements are safe enough. Claim 19: Basel III, and capital regulations as implemented in different jurisdictions, are based on careful scientific analysis of the cost and benefits of different levels of equity requirements, whereas the rough numbers of those who advocate much higher requirements cannot guide policy because they are not supported by scientific calibration. Claim 20: Because capital requirements should be adjusted to risk, it is essential to rely primarily on requirements that are based on assigning risk weights to assets. Claim 21: Instead of issuing more equity, banks should be required to issue debt or debt that converts to equity when a trigger is hit, so-called “contingent capital” or co-cos. Claim 22: Whereas equity is needed for banks as going concerns, banks in resolution need long￾term debt that can be bailed in. Total Loss-Absorbing Capacity (“TLAC”) in resolution must be large enough to permit a quick recovery. Claim 23: The Dodd-Frank Act in the US, or the newly adopted Banking Recovery and Resolution Directive (BRRD) and Single Resolution Mechanism in the European Union, have done away with the need to bail out banks; if a bank gets into trouble, the authority in charge of resolution will be able to resolve it without cost to taxpayers; there is therefore no need to increase equity requirements. Claim 24: If equity requirements are increased, banks will increase their “risk appetite,” which will make the system more dangerous. Claim 25: If equity requirements are increased, bank managers will be less disciplined. Claim 26: The best way to make banking safer is to require banks to put funds from deposits into reserves of central bank money or Treasury Bills (so-called narrow banking, also known as the Chicago Plan for 100% reserve banking). Such a shift will give us a stable financial system, and there would be less need to impose equity requirements. Claim 27: The financial system would be safe if banks are subject to a 100% reserve requirement so they can take no risk with depositors' money, while non-bank financial institutions are entirely prohibited from borrowing.

4 Claim 28: Tighter regulation of banks, and in particular higher equity requirements, are undesirable because they would cause activities to move to the unregulated shadow banking system. Claim 29: Since banking is a global business, it is important to maintain a “level playing field”. Therefore, banking regulation must be coordinated and harmonized worldwide. Claim 30: Stricter national regulation would harm “our” banks; instead we should be supporting them in global competition. Claim 31: The politics of banking makes effective regulation impossible, and therefore debating the merits of specific regulations such as equity requirement is “beside the point.”

5 Flawed Claims Debunked Flawed Claim 1: Capital is money that banks hold or set aside as cash reserve, like a rainy day fund; higher capital requirements keep money out of the economy. 3 What’s wrong with this claim? This statement is plainly false. As discussed in Chapters 1 and 6, capital in banking is a source of funding that can be used to make loans and other investments. This source of funding, elsewhere called equity, must be distinguished from debt, i.e., funds obtained by borrowing. Whereas banks typically fund less than 10% of their investments by equity, it is rare for any healthy non-financial company to have less than 25% in equity, and many successful companies borrow little or nothing, although there is no regulation that prevents them from borrowing as much as they would like (if they can find lenders). Flawed Claim 2: Requiring banks to hold cash reserves equal to 15% or more of their assets does not make them significantly safer, and therefore even such high capital requirement would not address the key problems in banking.4 What’s wrong with this claim? This claim rests on the same confusion between bank capital (equity) and cash reserves as Claim 1. Bank capital is not a cash reserve but a way of funding the bank. Capital requirements do not impose any restriction on what assets banks hold. They do not require banks to hold cash reserve. Since current requirements, and even the proposed Basel III reform, allow banks to have as little as 3% equity relative to their total assets, requiring 15%, or even 30% would make banks significantly safer. With equity levels considered minimal for healthy companies in the rest of the economy, banks would be able to absorb significantly more losses without becoming distressed or insolvent and without needing support, and, as we discuss in many writings, many distortions in the economy would be alleviated.5

Unlike equity requirements, reserve requirements are not as useful for maintaining the safety of banks unless they are very high. For example, if a bank has $97 billion in deposits and $3 billion in equity funding, cash reserve of $15 billion will not help it to survive if it loses $4 billion on its                                                              3 For example, Tim Pawlenty, president of the Financial Services Roundtable, is quoted in Wall Street Journal on July 20, 2015 (“Fed Lifts Capital Requirements for Banks,” by Ryan Tracy, Victoria McGrane and Justin Bear) claiming that “this rule will keep billions of dollars out of the economy.” In “How to solve the bank capital Goldilocks question,” CNN Money and Fortune, May 6, 2013, Cyrus Sanati falsely claims that capital requirements ask banks to “hold some cash on the sidelines.” The comparison of capital to “a rainy day fund” has also been used in Andrew Ross Sorkin, “Easing of Rules for Banks Acknowledges Reality” New York Times, January 7, 2013, and in Gretchen Morgenson, “Trying to Slam the Bailout Door,” New York Times, April 27, 2013. By contrast, “A Simple Rule to Stop the Next Financial Crisis,” Pat Ragnier, CNN Money, September 15, 2014, and “Bank Rules Debate, Capital Matters – and Words Too,” by Peter Coy, Bloomberg BusinessWeek, September 18, 2014 help clarify the terminology. 4 See for example, Cyrus Sanati, cited in footnote 3, who criticizes the higher capital requirements proposed by Senators Brown and Vitter and who, throughout the piece, falsely refers to the proposal as if it concerns cash reserves. 5 See Chapter 6, Admati et al (2013, Section 2).

6 loans and other investments. After the loss, it has $96 billion in assets and is insolvent, just as a homeowner is “under water” if the mortgage is larger than the value of the house. If instead the bank had $85 billion in deposits and $15 billion in equity, it would easily withstand the $4 billion loss and even a much larger loss without becoming distressed or insolvent. (However, see the discussion of Claims 23-24 regarding 100% reserve requirements.) Flawed Claim 3: The argument for requiring banks to have substantially more equity is only based on a theoretical result called the Modigliani-Miller theorem. This result does not apply in the real world because its assumptions are unrealistic.6 What’s wrong with this claim? Chapter 7 discusses the Modigliani-Miller theorem, which says that under some special conditions, a company’s mix of equity and debt funding does not affect the company’s overall value and funding costs, just like cutting a pizza into six slices instead of eight does not change the size of the pizza. The key insight of Modigliani and Miller, which holds universally, is that purely re-arranging how the risk taken by a corporation is divided among investor does not by itself change its funding costs.7 Other considerations may affect the funding costs, but they do not change our conclusions, as discussed in the context of Claims 4-13 below. Our argument for requiring much more equity is not in any way based on the presumption that the funding mix, in banking or elsewhere, is irrelevant. Our argument is based, as it should be, on a comparison of the costs and benefits to society of different funding mixes for banks. We argue, in particular, that there is a large cost, and no benefit to society, from having banks funded with as much debt as they can under current and proposed regulations allow. Flawed Claim 4: The key insights from corporate finance are not relevant for banks because the economics of funding for banks is entirely different from that of other companies.8 What’s wrong with this claim? Chapter 7 contains a section (pages 110-112) entitled: “The Big Question: Are Banks Special?” that directly takes on the claim “Modigliani-Miller does not apply to banks.” What is meant by this claim depends on whether “Modigliani-Miller” is considered as the “irrelevance” result or as an analytical approach. Whereas, as discussed in the context of Flawed Claim 3 above, the irrelevance result holds only under special assumptions,                                                              6 For example, Barclays Credit Research, “The Costs of a Safer Financial System,” March 25, 2013, Clearing House, “Vanquishing TBTF,” March 26, 2013, Oxford Economics, “Analyzing the impact of bank capital and liquidity regulations on US economic growth (A report prepared for The Clearing House), April 2013. 7 See also other references there, and Pfleiderer (2015). See also “Banks fail to convince crying foul over Basel reforms,” David Miles, Financial Times, November 23, 2010. 8 See, for example, Oxford Economics and Barclays Credit Research, both referenced in footnote 6, and “Safety in Numbers,” The Economist, April 11, 2013. DeAngelo and Stulz (2015), mis-characterize our arguments as relying only on Modigliani and Miller and proceed to develop a model of liquidity benefits from deposits in a model that assumes no uncertainty, which is hardly suited for discussing the notion of “liquidity.” The DeAngelo and Stulz analysis involves assumptions about pricing and the appropriation of consumer surplus by banks that are incompatible with market equilibrium, which they never actually study.

7 the analytical approach applies to all firms, including banks. Denying the relevance of the key insight of Modigliani and Miller is akin to denying the universal relevance of the laws of gravity in the presence of air frictions. The difference between the irrelevance result of Modigliani and Miller and the logic of Modigliani and Miller is easily seen when considering the role of liquidity benefits from bank deposits. It is widely acknowledged that such benefits invalidate the Modigliani and Miller irrelevance result, which presumes that participants care only about portfolio returns.9 However, this observation does not imply that it is efficient for banks to fund entirely by deposits, i.e. debt. If the liquidity of deposits depends on the bank’s not going into bankruptcy, then, with uncertainty, the bank needs equity to support the liquidity of its deposits by reducing the risk of bankruptcy. The loss of benefits from the liquidity of deposits represents a bankruptcy cost which according to the logic of Modigliani and Miller is part of the tradeoff determining the bank’s optimal funding mix – and moving the optimum in the direction of more equity funding.10 For bank equity and borrowing in wholesale markets and bond markets, the logic of Modigliani and Miller applies fully. Banks interact with the same investors that buy shares and bonds of other corporations. These investors value banks’ shares and bonds in the context of their overall portfolio and using the same criteria for all investments. The logic of how funding costs and the risks borne by different investors depend on the banks’ funding mix applies also to the borrowing by banks from other financial institutions.11 For large banks, this observation is important because typically more than half of their debt funding comes from markets rather than depositors. Importantly, like all other firms, banks can choose how much equity to use for funding and how much to borrow. And, like other firms, banks are more likely to become distressed or insolvent when they are highly indebted and have little equity. Moreover, the issues discussed in Chapter 3, entitled “The dark side of borrowing,” including the strong conflicts of interest between borrowers and creditors, and the distortions and inefficiencies of high indebtedness and particularly of distress and insolvency, apply to banks. Because of these distortions, the dynamics of leverage are characterized by an excessive growth of debt, which again exacerbates the distortions, as discussed in Admati et al (2015). As we show, once debt is in place, the funding mix is no longer determined by the choice that maximizes the total firm value, but rather chosen to benefit managers and shareholders even if the result destroys total firm value, in                                                              9 See Admati et al. (2013) and Chapter 10, as well as Gorton and Winton (2014), DeAngelo and Stulz (2015). 10 DeAngelo and Stulz (2015) miss out on this effect because they assume that there is no uncertainty. They also miss out on the possibility that, because of debt overhang effects, ongoing funding choices may not be value maximizing, see Admati et al. (2015). Gorton and Winton (2014) allow for this effect in principle but then impose special assumptions that eliminate it again, namely, either there are no assets outside of banks or equity is raised at a time when the incidence of risk is already known. 11 In some of the academic literature on banking, the statement “MM does not apply to banks” is used to postulate frictions that, under the assumptions of the models, might be addressed by borrowing, while conveniently ignoring the enormous frictions and collateral damage on the system that borrowing creates, which we discuss in Chapters 3, 6, 8 and 9. See also Pfleiderer (2015) and Admati and Hellwig (2013).

8 contrast to standard corporate finance results. Those who argue that banks are different and seek to justify the banks’ choice of funding mix as inevitable or efficient often neglect these distortions and inefficiencies, which can spill over to taxpayers and the public.12 Flawed Claim 5: Banks are special because they create money.13 What’s wrong with this claim? This claim rests on an abuse of the word “money.”14 The notion that banks “produce” or “create” money is based on the observation that people can easily transform deposits into cash and that they regard the funds they have in a bank deposit as being similar to cash and are able to use those funds for payments, such as by checks and credit cards.15 Monetary economists therefore refer to people’s total holdings of cash and of deposits in the economy as the amount of “money” in the economy.16 “Money creation” in the sense described above is related to banks’ holding so-called fractional reserves, i.e. keeping a fraction of the funds deposited with them as cash reserves and using the remainder for loans. As the banks’ borrowers use the funds they get to make payments, the recipients will keep parts of these payments in bank deposits. In this way, fractional reserve banking causes total deposits to be larger than the amount of central bank money deposited with the banks. The amount of “money” measured as the sum of deposits and cash in the economy is thus bigger than the amount of money that the central bank has issued. Putting demand deposits and cash into the same macroeconomic aggregate does not mean that they are literally the same. A critical difference is that deposits are a form of debt. 17 Banks are                                                              12 On the inefficiency of high leverage even from the private perspective of all banks' investors, see Admati et al (2015). 13 We have been confronted with this statement in various discussions of the book. 14 For a forceful criticism of this abuse of language, see Tobin (1967). 15 Some (for example, Gorton, 2010) have suggested that the use of short-term borrowing, for example through so￾called repos, is a modern-day form of deposits. (See also Cyrus Sanati, referenced in footnote 3.) Repos share with deposits the very short-term nature of the lenders’ claims. Unlike deposits, however, repo borrowing is not accompanied by provision of payment services. The repo lender, e.g., a money market mutual fund, might provide payment services to its own investors, but those services have nothing to do with the bank that acts as repo borrower. Repo borrowing takes place in wholesale markets with financial institutions acting as lenders. In these markets, as discussed in the context of Claim 4, the insights about the economics of funding that apply to all firms are fully relevant. The so-called shadow banking system, with money market mutual funds offering money-like claims and investing the funds they get in short-term claims on banks as well as other institutions, poses problems for monetary policy as well as prudential regulation and supervision. On the former, see Pozsar (2014), on the latter Gorton and Metrick (2010). 16 The value of this amount depends on how one draws the line between claims that are “money like” and claims that are not, for example whether one considers savings deposits or term deposits to be sufficiently similar to demand deposits to be included. Pozsar (2014) suggests that the amount of “money like” claims in non-bank institutions such as money market mutual funds should also be taken in. 17 One of the strangest statements in this context comes from John Stumpf, the CEO of Wells Fargo Bank, who reportedly said in an interview: “Because we have this substantial self-funding with consumer deposits we don’t have a lot of Debt.” (See Tom Braithwaite, “Wells Chief warns Fed over Debt proposal,” Financial Times, June 2, 2013. “Self-funding” ordinarily refers to equity and retained earnings. Deposits, by contrast, are a form of debt. It is false, indeed a contradiction in terms, to say that a bank that relies primarily on deposit funding does not have a lot of debt.

9 obliged to pay the depositor when he or she wants the money back. If a bank cannot repay depositors, there is clearly a problem. By contrast, cash, issued by a central bank, is nobody’s debt.18 (For a detailed discussion, see Chapter 10.) Flawed Claim 6: Bank deposits differ from other kinds of debt because banks create deposits by lending. What is wrong with this claim? This claim is often made in opposition to a “loanable funds” view of banks as intermediaries that collect deposits in order to fund their loans. Moreover, this “money creation through lending” is said to be the way money from the central bank gets into the economy.19 The claim rests on a confusion between stocks and flows. Indeed, if a commercial bank makes a loan to a nonfinancial firm or to a private household it provides its borrowers with a claim on a deposit account. Whereas this fact provides a link between the flow of new lending and the flow of new deposits, it is hardly relevant for the bank’s funding policy, which concerns the stocks of different kinds of debt and equity that it has outstanding, which must cover the stocks of claims on borrowers and other assets that the bank holds. A nonfinancial firm or household that receives a loan from a bank will typically use the associated claim on a deposit account for payments to third parties. The recipients of these payments may want to put some of the money they get into deposits, but they may instead prefer to move the money out of the banking system altogether, e.g., to a money market fund or a stock investment fund. 20 From the perspective of the individual bank, the fact that lending goes along with deposit creation does not change the fact that the bank owes its depositors the full amount they deposited. The key difference between deposits and other kinds of debt is not that deposits are “like money” or that deposits may be created by lending, but rather that the bank provides depositors with services such as payments through checks and credit cards or ATM machines that make funds available continuously. The demand for deposits depends on these services, as well as the interest that the bank may offer, and it may also depend on the risk of the bank becoming insolvent or defaulting.21                                                              18 Deposits with the central bank usually are claims to receive cash. Since the central bank can itself create this cash, these deposits do not involve serious obligations for the central bank 19 See McLeay et al (2014), and Jakab and Kumhof (2015). See also Thomas Mayer, “Lasst Bankpleiten zu!” (Allow banks to fail!), Frankfurter Allgemeine Sonntagszeitung, January 5, 2014, Martin Wolf, “Only the Ignorant Live in Fear of Inflation,” Financial Times April 10, 2014, and “Strip Private Banks of Their Power to Create Money”, Financial Times April 24, 2014. 20 Nor is it the case that lending by commercial banks is necessary for central bank money to get into the economy. Central bank lending to commercial banks and the latter's lending and deposit creation represent only one channel by which the central bank can inject money into the economy. Another way is for the central bank to buy securities such as government bonds or shares of private companies in the open market. The sellers of such shares might be private investors rather than banks, in which case bank lending plays no role in the central bank’s money creation at all. 21 Because depositors get returns through services just as well as, or instead of interest payments, the key insight of Modigliani and Miller is less important for deposits. As discussed in the context of Claims 4-5 and as explained in

10 The suggestion that bank lending is the only source of deposit creation is plainly false.22 Deposits are created when people bring cash to the bank, and they are destroyed when people withdraw cash. In this case, the reduction in deposits – like any reduction in funding – goes along with a reduction in the bank’s assets, i.e., a shortening of its balance sheet, but this reduction affects the bank’s cash reserves rather than its lending. The impact of such withdrawals on banks and entire banking systems are well known from the Great Depression or from the recent experience of Greece. In Greece in the spring and summer of 2015, depositors also were worried about the prospect that in the event of the country’s exit from the euro, their denomination of their deposits would be changed, whereas a stack of bills under a matrass would not be affected. Flawed Claim 7: Increasing equity requirements would reduce banks' ability to take people's deposits and issue short-term claims that are liquid and can be used like money. 23 What’s wrong with this claim? The claim falsely assumes that the amount of a bank’s equity is fixed and limited, and that none of the banks’ debt can be replaced with equity without interfering with “liquidity provision.” In fact, a bank can raise the amount of equity by retaining and reinvesting its earnings, or by issuing new shares, either in addition or instead of some of its debt. By increasing its equity, the bank could actually raise the amount of deposits it can take; if equity requirements are increased, adding equity would allow the bank to keep its deposits and other “liquid” debts unchanged. Relying on more equity would actually enhance a bank’s ability to provide useful liquidity because, with more equity, the bank’s debt is more trustworthy. Thus, contrary to the claim, the “liquidity” or “money-like” nature of deposits and other short-term bank debt is actually improved when the bank is less highly indebted and has more equity. By making the banks' deposits and other short-term debt safer, additional equity actually enhances the banks' ability to provide benefits to depositors without needing support from central banks or governments. In this context, however, the banks have flawed incentives, which lead them to borrow excessively. If the banks' owners and managers could firmly commit all their future funding decisions, they would take account of the fact that additional equity enhances the safety and the liquidity of their debt and makes the creditors willing to accept lower interest rates. As a matter of fact, however, such commitment is impossible. Over time, banks repeatedly take new funding                                                                                                                                                                                                 Chapter 7, however, this insight is essential for the other borrowing banks do, such as short-term borrowing from money market funds or hedge funds, and to their equity.. 22 This suggestion is made in McLean et al. (2014). 23 Barclays Credit Research, referenced in footnote 6, DeAngelo and Stulz (2015), The Economist, referenced in footnote 8, and Kling, “What Do Banks Do?” The American, February 26, 2013 warn of the reduction in bank deposits that, in their view, would be implied by higher equity requirements. Gorton (2012) refers to banks as “producers of debt” in the form of deposits and other short-term claims that people want because these debts are similar to money. Gorton views equity and investments as “inputs” for this debt “production.” There is actually no sense in which the bank’s equity is an input to its debt when both debt and equity entitle investors to payments from the bank, both being on the same side of the bank’s balance sheet. Indeed, it makes little sense to refer to debt promises the bank makes to its creditors as something that is “produced.”

11 decisions. In these decisions, the interest rates on previously-contracted debt are taken as given. Banks have no reason to take into account the fact that additional equity makes their previously-contracted debt safer whereas additional debt and the risky investments funded with this new debt make it less safe. Debt overhang, i.e., the existence of previously-contracted debts, may generate a ratchet effect that makes the bank's leverage increase whenever new needs or opportunities call for additional borrowing, whereas there is an aversion (on the part of the bank's owners, shareholders and managers) to reduce leverage because such a reduction would benefit incumbent debt holders. Because of this effect, the mix of debt and equity funding of banks that we see is likely to take insufficient account of the beneficial effects of additional equity for the safety and liquidity of deposits and other reforms of "money-like debt" of banks, in addition to not taking account of the effects of the risks to which their actions expose the rest of the financial system and the overall economy. The leverage ratchet effect is explores in detail in Admati et al (2015), and it is highly relevant for banks because of their already-high indebtedness and the passivity of their creditors, particularly depositors. The discussion above also suggests that the increased reliance of banks on short-term debt that we have seen in the past decade cannot be presumed to be beneficial for society or even privately for the banks. More likely, as we explain in our book and in other writings, this increase reflects the flawed incentives that banks' managers and shareholders have as a result of debt overhang.24 Flawed Claim 8: Increasing equity requirements would increase the funding costs of banks because investors require higher returns when investing in equity than when investing in debt.25 What’s wrong with this claim? First, as discussed in Chapter 7, it is fallacious to suggest that using more equity in the funding mix is more costly on the basis of the mere observation that the required return on equity is higher than the required return on debt. The required return on equity, debt, or any other security depends on the entire funding mix, and the required return on equity (as well as generally on other securities, including debt) will go down if the bank has more equity. As discussed in Chapter 9, and below in the context of Claim 13, a reason that total funding costs of banks might increase as a result of higher equity requirements is that with more equity banks would be less able to benefit from guarantees and subsidies, which come at the expense of taxpayers. For the policy debate, the relevant concern must be the cost and benefits to society of banks using different mixes of funding with different levels of equity. Because the fragility of the financial system is costly and harmful to society, a correct statement, contrary to the claim, is: “Increasing equity requirements would reduce the cost to society of having a fragile                                                              24 For more detail, see Admati et al (2013, Sections 4.2 and 4.3) and Admati et al (2015). The latter contains a detailed analysis of this effect as well as the method by which banks would choose to reduce leverage in response to leverage ratios requirements. This analysis and our recommendations in Chapter 11 of the book are relevant for making leverage regulation work.   25 See, for example, Oxford Economics, and Barclays Credit Research, (see footnote 6), The Economist (see footnote 8), and Elliott (2013).

12 and inefficient financial system where banks and other financial institutions borrow excessively, and thus it would be highly beneficial.” Flawed Claim 9: Increasing equity requirements would lower the banks’ return on equity (ROE) and thus make investors unwilling to invest in banks’ stocks. What’s wrong with this claim? As explained in Chapter 8, the first statement is false; when asset returns are low, the ROE is actually higher with more equity. Investors’ willingness to invest in banks’ stocks, or in the stocks of other firms, depends on whether they are properly compensated for the risk they take, not just on the stocks’ expected returns. If managers target specific ROE levels, they may actually harm shareholders by exposing them to risk without proper compensation. Moreover, when managers borrow excessively or take excessive risks, they harm creditors and taxpayers and endanger the public, which includes most of their shareholders. Flawed Claim 10: Increasing equity requirements would constrain banks so they must reduce lending.26 What’s wrong with this claim? As explained in Chapters 6 and 11, to comply with higher equity requirements, healthy banks can increase their equity levels by retaining their earnings or by selling new shares to investors. In either case, with more equity banks would have more funds, which can in turn be used to increase their lending. If increased equity requirements cause banks to reduce their lending, the reason is that they do not want to increase their equity. As explained in Chapters 3, 8, and 10 and in other writings, this phenomenon is due to the effect of overhanging debt and the conflicts of interest created by indebtedness which create a sort of addiction to borrowing that is reinforced and encouraged by government guarantees and by compensation structures in banking.27 Banks that are unable to raise equity at any price may well be insolvent and should be unwound, as discussed in Chapter 11. Banks’ lending decisions also depend on how attractive loans are relative to other investments. Many banks, including most of the large banks in the United States, are not even using all the funding they obtain from depositors to make loans.28 If banks do not make loans, therefore, the problem is not a lack of funds nor an inability to raise more funds for profitable loans, but rather                                                              26 See, for example, S&P, “Brown Vitter Bill: Game-Changing Regulation for U.S. banks, April 25, 2013. Elliott (2013) stresses that frictions in capital markets make it difficult or impossible for banks to raise new equity. As we discuss in Chapter 11, the arguments he gives that allude to information asymmetries are not applicable to new equity issues through rights offerings. 27 Admati et al. (2015) explore the leverage ratchet effect and explain why the effect is so important in banking. 28 See, for example, Elizabeth Dexheimer, “JPMorgan Leads U.S. Banks Lending Least Deposits in 5 Years,” Bloomberg, February 20, 2013. In the same story quotes a principal at Deloitte & Touche LLP, saying that new regulations that include “holding more capital to cushion losses” would impede lending. Quite obviously, especially in the context of the story (about the low ratio of loans to deposits), this statement is fallacious and misleading. This fact may not be obvious to readers because of the pervasive confusion between capital and cash reserves discussed in Flawed Claims 1-2.

13 the banks’ choices to focus on other investments instead.29 The risk-weighting system used in capital regulation, which we discuss in some detail in Chapter 11, also creates incentives for banks to invest in securities in the market rather than, for example, make business loans. Flawed Claim 11: Increasing equity requirements would be harmful for the economy because banks would be less willing to make loans.30 What’s wrong with this claim? This claim obscures the fact that credit crunches are primarily due to heavy indebtedness and financial distress, not from “too much equity.” More equity generally enables banks to increase their lending and to be able to continue to lend in downturns.31 As discussed in our response to the preceding claim and in Chapter 11, if banks choose to make fewer loans, the reason would most likely be because their overhanging debt makes issuing new shares unattractive or because they intensify their efforts at “risk weight management,” which, under the current system of capital regulation, induces a bias against lending and in favor of other investments. (See Flawed Claim 20 for more discussion of risk weights.) Controlling the transition to more equity by banning payouts to shareholders and specifying target levels of equity rather than ratios would mitigate any such effect. It is also false to presume that all lending is useful. Banks help the economy by making appropriate loans at appropriate interest rates that reflect the borrowers’ risks and the cost of funds. Some loans (such as, quite clearly some subprime mortgages prior to 2008) might actually be wasteful and inappropriate; such loans are usually the result of banks counting on someone else to bear the losses. Excessive lending can also result when there are too many banks with too much capacity; in this case, banks’ “gambling for survival” may offer cheap loans for a while, but their actions may expose the economy to increased risk of a major crisis later on. In fact, as already noted, credit crunch and reduced lending are due to the effect of debt overhang, which comes from excessive borrowing, not from having “too much equity.” The effect of higher capital requirements on lending is discussed in some detail in Admati et al (2013, Section 9).                                                              29 Under-investment is among the distortions and inefficiencies associated with heavy borrowing, again due to a “debt overhang” effect. This problem is explained in Chapter 3. 30 In addition to Barclays Credit Research, Oxford Economics (see footnotes 6), and Elliott (2013). The Clearing House, referenced in footnote 6, and S&P (see footnote 26), also warn that higher equity requirements would reduce the supply of credit. 31 In the same spirit, Mervyn King, the outgoing governor of the Bank of England, recently said: “Those who argue that requiring higher levels of capital will necessarily restrict lending are wrong. The reverse is true. It is insufficient capital that restricts lending. That is why some of our weaker banks are shrinking their balance sheets. Capital supports lending and provides resilience. And, without a resilient banking system, it will be difficult to sustain a recovery.” (See King (2013).) .) Kapan and Minoiu (2013) show that “banks with strong balance sheets were better able to maintain lending during the crisis,” and suggest that “strong bank balance sheets are key for the recovery of credit following crises.” Cole (2013) shows that bank lending to businesses suffered when banks incurred losses and that the Troubled Asset Relief Program (TARP), which did not alleviate the banks' indebtedness, did not result in improved lending.

14 Flawed Claim 12: Higher equity requirements would restrict banks’ ability to provide market￾making services, harm market liquidity, and prevent banks from stabilizing volatile stock markets.32 What’s wrong with this claim? There is no automatic connection between equity requirements and the ability of banks to provide market-making services, to enhance market liquidity, or to stabilize stock market. doFor example, banks avoiding a safe investment such as in US Treasury bond, in favor of a risky and more rewarding trade. Such behavior generally harms creditors and taxpayers. If banks merely act on commission for their customers, their own accounts are not affected at all. If banks act as counterparties, buying securities that the customers want to sell or selling securities that the customers want to buy, the question is how these transactions fit into the bank’s own asset management and portfolio decisions. As with the Flawed Claims 10 and 11, about bank lending, such questions of portfolio choice do not depend on the bank’s funding mix. To comply with higher equity requirements, banks retain their earnings and raise more equity. With ample equity, it is moreover likely that asset choices are undistorted by excessive incentives to take risks. From the customers’ perspective, banks’ professed desire to provide market making services can be a mixed blessing. In many instances in the past, banks have used their customers’ dependence on such services in order to take advantage of the information provided by customers’ orders, using practices such as front-running or dual-capacity trading to speculate on the basis of privileged information about their customers’ orders.33 Whereas banks claim that their services improve market liquidity, such practices, which are almost impossible to prevent, actually harm the customers’ confidence and the liquidity of the markets. Market liquidity captures the ability and ease of converting financial securities to cash through trading in markets and the price at which securities can be bought and sold. Liquidity is determined by the balance of reasons for trading of various market participants, namely the availability of buyers and sellers at a given time and price, the trading mechanism that determines the market price (for example, how buyers and sellers find each other, whether an intermediary or an exchange is involved, etc.), and, importantly, on the information that participants have about the value of the security, which may differ across participants. Liquidity                                                              32 For example, “Bank Capital Needs Seen Soaring on Basel Market-Risk Review,” John Glover and Boris Groendahl, Bloomberg, October 20, 2015, quotes Mark Gheerbrant from the International Swaps and Derivatives Association (ISDA) saying “We’re concerned about the impact [increased capital requirements] will have on market liquidity”. Jürgen FItschen, Co-CEO of Deutsche Bank claimed that, as saw recently, banks will be less able to counter adverse market moves, with reference to the stock market declines in the summer of 2015 (see “Wenn e suns zu gut geht, machen wir Fehler” – “If we are doing too well, we make mistakes,” Handelsblatt, September 3, 2015,). 33 See, for example, Pagano and Roell (1990, 1993) and Roell (1990). The debate on high-frequency trading involves similar issues. Some of the episodes in Lewis (2014) illustrate the problem and the potentially large social costs involved.

15 can be reduced, or even break down, if some participants have much better information than others, creating so-called adverse selection (similar to the market for used car). Whether some financial institutions must use more equity funding for their trading does not bear directly on any of these considerations. 34 There is also no automatic connection between equity requirements and the positions banks take in stock markets, except perhaps that stock market investments might involve higher risk weights than other kinds of investments. Solvent banks can always raise additional equity if their portfolio decisions require it. Higher risk weights for stock market investments – or prohibitions of stock market investments under the Glass-Steagall Act or the Volcker Role in the United States) – reflect the assessment that such investments may be too dangerous for banks. The banker’s promise to try to prevent stock prices from falling, if actually not empty, should raise concerns about the risks the banks are taking. When banks tried to stop and reverse a falling stock market on October 24, 1929, it only took four days for them to realize the futility of the effort and the size of the losses they had incurred.35 In this context, it is useful to note that the worldwide decline in stock market values after the Lehman Brothers bankruptcy amounted to some $20 trillion, three or four times the decline after the burst of the tech bubble in the early 2000s. This decline was greatly exacerbated by banks’ scrambling for cash and selling assets as money markets on which they had relied for funding ceased to function. Lack of equity to absorb losses from “toxic” assets was one reason so many banks were mistrusted and were unable to roll over their short-term funding. Flawed Claim 13: Higher equity requirements are undesirable because they would prevent banks from taking advantage of government subsidies and thus force them to charge higher interest on loans.36 What’s wrong with this claim? Whereas deposit insurance is useful for preventing inefficient bank runs, it is often underpriced for individual banks, and it has the undesirable impact of enabling and encouraging banks to take risk and to “economize” on equity. Underpriced explicit or implicit guarantees to any form of bank borrowing make bank funding artificially cheap and create a distortion in the economy. By rewarding debt and penalizing equity funding the subsidies are socially harmful, especially at the very high levels of debt the banks choose. Even if all the subsidies are passed to banks' customers in the form of cheaper loans, they contribute the financial system’s being inefficient, bloated and fragile, and they distort competition and the allocation of resources in the economy.                                                              34 Liquidity requirements, which mandate the holdings of assets within specific set may have more of a direct income on the willingness of institutions to trade in, or make certain markets. 35 See Galbraith (1954, pp. 105-120). 36 See, for example, Oxford Economics, referenced in footnotes 6, and Elliott, referenced in footnote 17. William Isaac, in “Better than Brown-Vitter: Make Banks Issue Long-Term Debt,” American Banker, June 4, 2013 warns that higher equity requirements on the largest banks would cause them “to decrease their lending dramatically and/or increase significantly the price of loans.”

16 There are two types of subsidies banks receive when they borrow but not when they use equity funding. First, the tax code in most countries gives debt a tax advantage relative to equity for all corporations. Despite this tax treatment, and even with no regulation of their funding, no healthy corporation maintains as little equity as banks. The tax code has no economic rational and it is highly distortive. Whereas the effect can be neutralized, there is still no social cost if banks pay more taxes. (See Admati et al (2013, Section 4.1).) Second, explicit guarantees through deposit insurance for which banks often do not pay the appropriate economic costs, and implicit guarantees that allow banks to borrow at terms that are more favorable than their indebtedness and the risks they take would normally imply, encourage and subsidize excessive borrowing.37 Measuring the size of subsidy is difficult because it amounts to an underpriced insurance contract whose value changes with the likelihood and extent to which it will be needed. In fact, there is reason to believe even many academic studies under-estimate the subsidies.38 Despite the overwhelming evidence that the subsidies are substantial, large banks deny the existence of subsidies, while claiming that their cost of funding would increase with more equity. These claims are inconsistent with one another. Requirements that banks use much more equity do not impose a cost to society; rather, they attempt to correct distortions and reduce excessive subsidies. If it is deemed desirable to subsidize specific loans or any other activities, subsidies should be given directly to the intended recipient, for example by attaching the subsidies to specific loans. Blanket subsidies to bank borrowing, by contrast, provide banks with below-market funding that they can use at their discretion. The cheap funds may not actually go to the loans that the economy needs, and instead                                                              37 There is broad agreement that the subsidies are substantial. For example, see Chapter 3 of IMF 2014 Financial Stability Report, yet in documents such as, “Measuring the TBTF effect on bond pricing,” by Goldman Sachs Global Markets Institute, May 22, 2013, large banks argue that large banks do not benefit from a too-big-to-fail effect on their funding costs. There are a number of critical flaws in the Goldman Sachs analysis, and most are discussed in Mark Whitehouse “Too-Big-To-Fail Myths, Goldman Sachs Edition,” Bloomberg View, May 28, 2013. (See also Christopher Cole, “Goldman's TBTF Study Used Flawed Data to Reach Flawed Conclusions,” American Banker, May 30, 2013.) First, it compares interest rates on bonds of large banks and small banks without adjusting for differences in the risk creditors are supposedly exposed to. As discussed by Brando et al (2013), however, too￾big-to-fail banks tend to take more risks in their investments than smaller banks; unless the implicit guarantee is perfect, this would raise the interest TBTF banks have to pay. (Frank Partnoy and Jesse Eisinger, in “What is Inside America’s Banks,” The Atlantic, January 3, 2013 also shows banks’ disclosures make it difficult for investors to assess the risk.) Second, the observation that creditors suffer more in failures of small banks relative to those of large itself reflects too-big-to-fail policies, including support from the Federal Reserve that has provided ample and cheap funding to banks since 2008. The GS paper dismisses findings of a large literature (some of which is also cited in Chapter 9) without engaging on substance, including academic studies that conclude that the value of the subsidies is in the tens of billions of dollars and particularly large in downturns. Many other industry-sponsored studies also fail to correct properly for the funding mix and other parameters of the bank borrowing that would affect the risks that their long term creditors would be exposed to, relative to those of other companies that do not have access to safety nets. 38 See Stefan Nagel, “Too Big to Fail is Larger than You Think,” Bloomberg View, March 2, 2014. Given the opacity and complex structure of the liabilities of the largest banks, it is possible that without any guarantees, the cost of unsecured borrowing to these banks would be prohibitive. Of course, among the reasons banks are able to borrow as much using collaterals is that deposits are unsecured, and at least some assets purchased with deposits can be used as collateral for additional borrowing.

17 the borrowing itself makes banks more fragile, exposes the economy to substantial risks, and distorts banks' investment decisions, giving them incentives to take excessive risk in their investments or to under-invest in relatively safe but worthy loans because bankers do not find them to have enough upside.39 For more on these issues, see Chapter 9, entitled “Sweet Subsidies,” which discusses harmful effect of guarantees and subsidies, Admati (2014b), and Chapters 12 and 13. The critical distinction between private costs to the banks and social costs to society is discussed in more detail in Admati et al, (2013, Section 4). If banks' funding costs (or any costs to banks' shareholders) are increased as a result of them being less able to take advantage of subsidies, the impact is entirely private. The cost and the harm of excessive indebtedness in banking is borne by the broader public without producing any corresponding benefit. Nevertheless, subsidizing banks through implicit guarantees is attractive for policymakers, because it does not show on budgets as it is given, thus appearing costless. In fact, the costs to society of providing banks with outsized and highly distortive subsidies are large, and equity requirements that reduce these subsidies and correct the distortions are thus highly beneficial. Flawed Claim 14: Historically, banks have never had as much as 30% equity; requiring as much equity would therefore harm the business of banking.40 What’s wrong with this claim? The statement is false. First, references provided in our book (particularly in notes 20-27, pp. 242-243) support the claim that going back more than a century to the period before bank owners and shareholders could rely on creditors, central banks, or governments to pay their creditors, it was common for banks to have as much as 50% equity. Second, arguments based on history presume that circumstances are similar. However, since the 1970s (uninsurable) macroeconomic risks have become much larger than they had been in the preceding decades. More importantly, financial institutions worldwide have become much more interconnected; this has greatly increased systemic from contagion. In some parts of the business also competition has become much more intense; this has reduced the ability of banks to rely on margins to provide buffers against shocks. Our proposed leverage ratios do not stand on any historical figures, but are rather based on the economic arguments and observations of leverage in other, unregulated industries and on considerations of the social cost of banks' leverage. As indicated in the discussion of Flawed Claims 4-9, the economics of high leverage is not fundamentally different for banks even if some                                                              39 For example, Levitin (2014) questions our statements that there is no social cost in reducing distortive subsidies, missing the distinction between social and private costs that is explained in detail in Admati et al (2013, Section 4). Matt Yglesias, in “Banks Borrow Too Much,” Slate, March 7, 2013, expresses concerns regarding the potential cost of withdrawing the banks’ subsidies. However, in his subsequent blog post entitled “How I Learned to Stop Worrying and Love Higher Capital Requirements,” March 8, 2013, he states that in our book we “in many ways end up underselling the power of [our] idea,” emphasizing that, as we explain in Admati et al (2013, Sections 2 and 9), not only would more equity make banks safer, but it will also make their lending and investment decisions more appropriate and better for society. 40 Calomiris (2013) and Levitin (2014).

18 of banks' debt is useful for providing liquidity.41 Quite clearly, the bankruptcy of Lehman Brothers had significant collateral damage. As Admati et al (2013, 2015) explain, markets may allow leverage to get socially, and even privately, excessive. Requiring investment banks, which can scale up risk and become systemic, to have 30% equity corrects this situation and produces substantial social benefits with minimal if any relevant cost. We are sometimes asked why we do not go to 100% equity. The reason is precisely that deposits do provide benefits (beyond providing funding for the banks) that are not captured by standard corporate finance arguments. However, for most large banks today, deposits amount to less than half of their funding.42 The 30% ratio we propose is roughly what banks themselves impose on financial institutions, such as hedge funds or REITs, to which they lend. Flawed Claim 15: There is not enough equity around for banks to be funding with 30% equity.43 What’s wrong with this claim? As explained in the context of Claim 1, equity is not a cash reserve but a financial claim that banks can issue to obtain funding for their investments. Contrary to this claim, higher equity funding for banks does not require new savings and new inflows into capital markets. If a bank issues more equity and uses the funds it obtains to buy listed securities, capital markets will adjust so that investors who have sold the other securities will hold additional bank shares because the bank’s returns would partly reflect the returns on those other securities. No new savings and no new inflows of funds into capital markets are required. To the extent that all assets in the economy are held by, and all risks are borne ultimately by end investors and taxpayers, the effect of a reshuffling of financial claims to make sure more equity funds banks' investments would generate less distorted, more appropriately priced investments in the economy.44 As discussed below, the size of individual firms and of a sector in the economy should be determined in undistorted markets. If the banking sector has access capacity, it is best for the economy if unhealthy institutions that can survive only because of supports exit rather than persist. Flawed Claim 16: Because banks cannot raise equity, they will have to shrink if equity requirements are increased, and this will be bad for the economy.                                                              41 Levitin (2014) also argues that the market does not demand 20% or 30% of small banks that can fail or of investment banks. But small banks have insured deposits who don't bear deadweight bankruptcy costs, and Lehman Brothers' creditors may have well believed that they would be paid in full, as were the creditors of Bear Stearns even though they were not explicitly insured. 42 See Advisory Scientific Committee (2014), as well as Chapter 6. 43 For example, Elliott (2013) stresses that frictions in capital markets make it difficult or impossible for banks to raise new equity. As we discuss in Chapter 11, and in Admati et al (2015), the arguments he gives that allude to information asymmetries are not applicable to retention of earnings or to new equity issues through rights offerings 44 A more detailed discussion of this argument is offered in Sections 7 and 9 of Admati et al (2013). At current levels of indebtedness, individual institutions, and the banking sector as a whole, are likely to be inefficiently bloated due to excessive subsidies. See also the discussion of Claim 17 and Admati (2016).

19 What’s wrong with this claim? As we discuss in Chapter 11, solvent banks can always raise equity by selling additional shares, to existing shareholders through rights offerings or to new shareholders in the market. If a bank cannot raise equity at any price, the bank is likely to be insolvent.45 The existence of nonviable banks that cannot raise equity may reflect excess capacity in banking. (Excess capacity appears to be a serious problem in some countries and maybe globally at this time.) In this case, some downsizing of the industry would benefit the economy, contrary to the claim. The remaining banks would be viable and would have fewer incentives to gamble at the expense of their creditors, the taxpayers and the economy. Flawed Claim 17: Increasing equity requirements would harm economic growth.46 What’s wrong with this claim? Those who make this sweeping assertion do not typically provide a coherent explanation for why increased equity requirements, which amount to a reshuffling of financial claims in the economy, would have a harmful effect on growth. They also neglect the fact that the worst downturn in economic growth occurred as a result of the actions taken by highly indebted banks and other financial institutions, which led to the financial crisis in the last quarter of 2008. One reason for the severity of this crisis was the lack of equity in banks, which made banks vulnerable to the decline in US real estate markets, defaults on subprime mortgages and the collapse of the markets for asset-backed securities. Growth can suffer also when risk weights, discussed in the context of Claim 20 and in Chapter 11) distort investments away from worthy business loans and towards government lending and traded assets. Crowding out of private borrowing by government borrowing can have substantial negative effects on economic growth, as seen in southern Europe in the decades before 1990.47 Reference to the impact of higher equity requirements on bank lending ignores the fact that it is overhanging debt, and not excessive equity that lead to credit crunches, as discussed above in the context of Claim 11. In fact, banks with more equity to absorb losses without becoming distressed would be more able to sustain lending in a subsequent economic downturn, which would have positive effects for investment and the economy. Growth, as seen for example in Iceland and Ireland, can be temporary and illusionary when it reflects a boom that is followed by bust. As we discuss in Chapter 11, if the transition to a system with more equity funding for banks and other institutions is handled properly, there would be no negative consequences to making the financial system less indebted and thus safer and less distorted.                                                              45 For details of the argument, see Admati et al. (2013, 2015) 46 See for example Oxford Economics, referenced in footnote 5. Levitin (2014, p. 2036) complains that we have not dealt with this claim in the book even as he does not explain why the claim should be true (except that bankers such as Josef Ackermann have asserted it to be true). In the book and elsewhere (including in the current document) we have argued that whatever justification (if any) is given to this claim, it is invalid or misleading. 47 See, for example, the essays by Caminal et al., and Borges in Dermine (1990)

20 Flawed Claim 18: Basel III is already very tough, doubling or tripling previous requirements; banks that comply with Basel III requirements are safe enough.48 What’s wrong with this claim? As we discuss in Chapter 11 (on the basis of the arguments of previous chapters), these statements use a false benchmark for the desired and feasible equity levels. Basel III still allows banks to fund up to 97% of the assets on their balance sheets by borrowing, just as Lehman Brothers did. As discussed below regarding Claim 19, the numbers in Basel III are not based on sound analysis, and the papers justifying them are fundamentally flawed. Stress tests have also based on flawed approaches; they have been much derided when banks that the stress tests said were safe became insolvent only a few months afterwards.49 Moreover, the measurements of so-called bank capital often refer to accounting ratios of accounting measures of equity relative to risk-weighted assets, which has proven very poor for predicting banks' ability to withstand losses. Moreover, the regulations often rely on debt-like alternatives to equity, which are problematic. (See the discussion of Claims 21-23 below.) Flawed Claim 19: Basel III and capital regulation as implemented in different jurisdictions, are based on careful scientific analysis of the cost and benefits of different levels of equity requirements, whereas the rough numbers of those who advocate much higher requirements cannot guide policy because they are not supported by scientific calibration.50 What’s wrong with this claim? Basel III, and regulations as implemented, appear to be the result of a political process much more than of valid scientific analysis. As we discuss in Chapter 11 and elsewhere, the studies that support the Basel III rules are based on flawed models and their quantitative results are meaningless. For example, they assume that the required return on equity is independent of risk; one paper purports to derive the “optimality” of Basel III without even considering the costs that bank failures can impose on the rest of the financial system and the economy.51 The “scientific” papers that discuss costs and benefits of different capital requirements also ignore the distinction between private and social costs, the distortions in investments associated with high leverage, and the problems with risk weights, discussed below.52                                                              48 Claims that the requirements are tough and that banks are stronger now are frequently made by regulators, bankers and others. For example, Tom Braithwaite, in “Quest for Profits can Make Banks Safer,” Financial Times, February 18, 2013, suggests that the “lust for improved ROE” is a helpful, ignoring the possibility that a lust for return often involves taking risks and borrowing inefficiently, including to get around regulations based on risk weights. 49 We discuss stress tests in Chapter 11. See also Vestergaard and Retana (2013) and Dowd (2015) for critiques. 50 Claims that the requirements are tough and based on “science” are frequently made by regulators, bankers and others. For example, in a November 19, 2013 interview to Die Welt Lloyd Blankfein, CEO of Goldman Sachs, said: “The new capital adequacy regulations under Basel III are the results of a long and meticulous process.” 51 Admati et al (2013) discusses some of flaws in papers produced by the Basel Committee on Bank Supervision and other authors. 52 On the attempts to provide estimates, see also Admati and Hellwig (2013), Admati (2014a). A recent paper, Brooke et al (2015) still includes a flawed analysis of the tradeoffs. For a discussion of some of the flaws, see Admati (2016).

21 The fact that studies end up with precise numbers for “optimal” capital regulation is irrelevant if the foundations of the studies are shaky. We are not aware of any theory or model that would provide appropriate estimates of the costs and benefits to society associated with different funding mixes for banks. Despite this, we are confident in asserting that equity levels of three percent of total assets, as admitted by Basel III, are unsafe, and that a significant increase will substantially improve the health and safety of the financial system. Low levels of equity expose the banks and the economy to unnecessary risk. Allowing banks to rely as much on subsidized borrowing distorts and harms the economy. Countering the banks’ incentives to choose unsafe levels by effective regulation is essential, because, as discussed in Admati et al (2013, 2015), markets fail to produce an efficient outcome both privately for all of banks’ investors and socially. A significant challenge in specifying any specific capital ratio has to do with setting the appropriate numerator and denominator, which have to do with valuations of the relevant assets and liabilities. Accounting conventions matter greatly, including how to deal with off-balance￾sheet exposures and derivatives. As explained in Chapter 11, the key to capital regulation is high requirements for genuine, loss absorbing equity, and prompt intervention by regulators if equity is depleted through losses.53 Flawed Claim 20: Because capital requirements should be adjusted to risk, it is essential to rely primarily on requirements that are based on assigning risk weights to assets.54 What’s wrong with this claim? As we discuss in Chapter 11, the system of risk weights that we currently have has more to do with politics and tradition than with science. In fact, the Basel rules negate important sources of risk altogether: Risks from sovereign debt that is funded in the currency of the country in question, risks of changes in funding conditions for medium or long￾term loans, risks from the possibility that borrowers might default simultaneously because their default risks are correlated. Risk from sovereign debt that is funded in the currency of the country was in evidence in the Greek default in 2012. Funding risk for long-term loans was a key factor in the S&L crisis in the 1980s. Correlated borrower defaults were a major factor in the subprime mortgage crisis of 2006-2009. Even if the politics of the regulation could be dealt with, attempts to improve risk weighting are limited by a lack of data and by the never-ending changes in the risks and correlations. In practice, the system of risk weights allows banks to be extremely highly indebted, masks important risks, and adds to the interconnectedness of the system. Whereas proponents of the system argue that it is important to require banks to have more equity funding when their assets are more risky, in fact the system allows banks to get away with much less equity funding when                                                              53 On accounting issues and ways banks can manipulate them through securitization and derivatives, see Kerr (2011). For more on “numbers” see Admati (2014b, 2016). 54 For example, Tom Braithwaite (referenced in footnote 25) praises the Basel risk weights system for controlling banks’ risks. Most regulators appear to take it for granted that risk weights are essential, and the Federal Reserve has proposed to adopt Basel III, including the use of risk weights, for all US banks.

22 they say that their assets are less risky. A uniform ratio of required equity to total assets would provide a bound on the banks’ leverage. By contrast, because some risk weights are (near) zero, the risk-weighting system allows very high leverage. Thus, banks could take large positions in assets with (close to) zero risk weights, such as Greek sovereign debt or AAA-rated toxic securities, and fund them almost entirely with debt and with hardly any equity. The system also distorts banks investment decisions, typically against business lending, and is highly manipulable by the banks.55 The ability of banks to “economize on equity” is enhanced by their ability to use their own models to assess risks. The scope for manipulation they have is largest for assets in the trading book, which is why they were keen to put mortgage-backed securities and the like into the trading book, subject to mark-to-market accounting rules. Most of the losses in 2007-2009 were incurred on assets in the trading book, where equity often was as low as 1 percent of investments.56 Credit risk on assets in the bank book, i.e., assets that banks claimed they intended to hold to maturity, played less of a role in the crisis (except for sovereign exposures in the euro crisis). The changes in regulation (“Basel II”) that allow banks to use their own models to assess credit risk were only being introduced when the crisis unfolded. However recent empirical research has shown that the use of model-based internal ratings to assess credit risk and determine risk weights for capital regulation has gone along with a significant deterioration in the quality of these assessments: for comparable borrowers, internal ratings are better and actual risk incidence is worse than under the previously used “standard approach.”57 The Basel Committee has recently recognized that that there is a problem with risk weights. However, its focus is on heterogeneity of risk assessments across banks as evidence of problems in assessment procedures, rather than the scope for manipulation that the model-based approach offers altogether. The Committee tries to impose more control and more homogeneity on risk assessment procedures without however questioning the approach as such.58 The risk weighting system is particularly harmful because equity levels are as low as they are, thus the incentives to increase leverage and take excessive risk are particularly intense, leading to a focus on manipulating risk weights that undermines the regulations and exacerbates systemic risk.                                                              55 See further discussion in Chapter 11. The report on JP Morgan Chase “London Whale” loss by the Senate Committee on Investigation specifically refers to attempts to manipulate models, and points to poor risk controls. See, for example, David Henry and Lauren Tara LaCapra, “JPMorgan and other banks tinker with risk models,” Reuters, March 18, 2013, and Floyd Norris, “Masked by Gibberish, the Risks Run Amok,” New York Times, March 21, 2013. On the flaws in the Basel approach, see also Haldane (2012, 2013), Hoenig (2013), Joe Rizzi, “Risk-Based Capital: The Good, the Bad and Mostly the Ugly,” American Banker, May 20, 2013, William Isaac, referenced in Footnote 37, and Robert T. Taylor, “Basel Rules Echo Missteps of S&L Era,” American Banker, June 19, 2013. See also a related discussion in Dowd (2015) in the context of stress test. 56 FSA (2010). 57 See Behn et al. (2014). 58 See Basel Committtee on Bank Supervision BCBS (2015)

23 Flawed Claim 21: Instead of issuing more equity, banks should be required to issue debt that converts to equity when a trigger is hit, so-called “contingent capital” or co-cos.59 What’s wrong with this claim? As we explain in Chapter 11 (pp. 187-188), in a section entitled “Anything but Equity,” and in Admati et al (2013, Section 8), the various proposals to use hybrids between debt and equity as a way of forcing investors rather than taxpayers to bear losses offer no advantages, and in fact have important disadvantages, relative to equity. First, like other debt, they raise the specter of domino effects or near the triggers where debt converts to equity (or is written down, depending on what the contract says). If the institutions that hold the co-cos are systemic, the consequences of a conversion to equity can be dramatic, and fear of these consequences might motivate a bailout. Indeed, in 2008-2009, holders of long-term debt and other hybrid securities meant to absorb losses as Tier 2 capital were paid even as banks were bailed out with taxpayer funds. Second, when conversion is imminent, the strategic behavior of market participants can induce dramatic changes in prices of equity and/or co-cos. Thus, co-cos do not provide reliable loss absorption and can create instability in a crisis. Third, as long as they have not been converted to equity, co-cos and other debt-like claims add distortions to banks’ lending decisions by exacerbating the effect of debt overhang and contributing to credit reductions in downturns. There is no sense in which having banks rely on these hybrid securities is “cheaper” or better for society than relying on equity. For the purpose of regulation, using equity simply dominates these alternatives. Those who propose such alternatives as a substitute for equity have yet to give a valid reason for their proposal that is relevant for policy considerations.60 Flawed Claim 22: Whereas equity is needed for banks as going concerns, banks in resolution need long-term debt that can be bailed in. Total Loss-Absorbing Capacity (“TLAC”) in resolution must be large enough to permit a quick recovery.61 What’s wrong with this claim? The suggestion that debt that serve as TLAC (or, as the European Banking Recovery and Resolution Directive calls it, bail-in-able debt) can do something that equity cannot do is false and misleading. Obviously, once a bank is insolvent,                                                              59 See, for example, Calomiris (2014). Proposals to use co-cos instead of equity have been implemented in Switzerland and have been discussed in the UK (see UK Independent Commission on Banking) and the European Union (see Liikanen Report). A variation on the concept is Equity Recourse Notes (ERNs) proposed by Bulow and Klemperer (2014), which amount to debt whose coupon payments are made in equity when a trigger is hit. 60 As discussed in the context of Claim 13, compromising financial stability in order to give tax subsidies to inefficient funding by banks makes no sense. (Because they can force conversion to equity and do not confer creditors’ rights on their holders, co-cos do not qualify as debt under the US tax code. and thus do not have the tax advantage over equity in the US that they appear to have in Europe.) On the claim that long-term debt provides better discipline than equity, see the discussion of Claim 25 below. Co-cos and ERNs that they are meant to convert some debt to equity ahead of insolvency and failure are better than debt claims that can only suffer losses within a “bail-in” process or in a resolution or bankruptcy (which are discussed in the context of Flawed Claim 23 below). And they are obviously a less fragile funding source than short-term debt that is subject to runs. 61 See FSB (2015).

24 there is no equity left and thus any losses must be borne by someone else, which must be some debt holders if a bailout is to be avoided. However, the point of insolvency itself is immediately affected by how much equity there is to begin with. The more equity there is, the more losses it can absorb so as to avoid insolvency and entry into resolution in the first place. The total loss absorbing capacity of equity and bail-in-able debt is not increased when equity is replaced by bail-in-able debt. To the contrary, if the authorities end up being unwilling to impose losses on debt holders, a replacement of equity by bail-in-able debt reduces loss absorption capacity. The arguments in the discussion of Flawed Claim 21 concerning co-cos and the likelihood that holders of co-cos might be bailed out after all apply equally to bail-in-able debt or TLAC. Legally, the holders of TLAC have stronger claims than those of the holders of hybrid (convertible) debt considered as regulatory Tier 2 capital before the crisis but in 2008, yet even those weaker claims were bailed out routinely and did not absorb losses. The one exception to this rule, Washington Mutual, was highly disputed inside the US Government, and the systemic effects from the bail-in of unsecured senior debt holders of Washington Mutual has convinced many that, in a systemic crisis, such bail-ins are to be avoided. These considerations are bound to be brought back if there is a question of bailing in unsecured senior debt in a situation of systemic stress. Holders of bail￾in-able debt may also be small savers who have not realized that they might be called upon to absorb the banks’ losses, as happened in Spain and more recently in Italy, thus causing a political problem if losses are large or many institutions fail.62 In sum, this flawed claims neglects the fact that equity absorbs losses more reliably than any other form of “loss absorbing capital,” and that it is best to use used equity instead of any poor alternatives. It is precisely the virtue of equity that it absorbs losses without anyone triggering a formal resolution procedure. Systemic effects from the triggering of such a procedure may well prevent the procedure from being triggered at all, in which case any notion of loss absorption by certain debt instruments is moot.63 Flawed Claim 23: The Dodd-Frank Act in the US, or the newly adopted Banking Recovery and Resolution Directive (BRRD) and Single Resolution Mechanism in the European Union, have done away with the need to bail out banks; if a bank gets into trouble, the authority in charge of resolution will be able to resolve it without cost to taxpayers; there is therefore no need to increase equity requirements.64                                                              62 See “Italy bank rescues spark bail-in debate as anger at Renzi grows,” James Politi, Financial Times, December 22, 2015. 63 Persaud (2015) calls the bail-in concept “fool’s gold” as a solution to the too-big-to-fail problem. 64 See, for example, presentation by the Clearing House to the Board of Governors of the Federal Reserve regarding Title II of Dodd Frank Act on February 13, 2013, and their March 26, 2013“Vanquishing TBTF.” See also William Isaac, referenced in Footnote 37. This claim is the basis for proposals by the Federal Reserve to force bank holding companies to use more long-term debt (see, e.g.,, Governor Daniel Tarullo testimony to Senate Committee on Banking, Housing and Urban Affairs, February 6, 2014), and similar discussions by the Financial Stability Board

25 What’s wrong with this claim? As we discuss in Chapters 5 and 9, this claim ignores a number of critical points and is not credible.65 First, to minimize the economic disruptions from having banks go into resolution, it may be necessary to maintain some important operations at least temporarily. This requires funding. Under the Dodd-Frank Act, such funding might be obtained by borrowing from the government; such borrowing puts the taxpayer at risk.66 Second, whereas both the Dodd-Frank Act in the US and the BRRD in the EU rely on industry levies and on creditor bail-ins to absorb losses, in a crisis, when many banks may be weak at the same time and the financial system is at risk, the industry as a whole or the banks’ creditors (which may be other financial institutions) may be too weak to perform this role. Even if the charges are spread over time, the burden of obligations they impose may be so great that the institutions involved become incapable of functioning. These concerns arise even if the debt in question is long-term or, as in Claims 21-22, subject to contingent conversion clauses. If the banks were required to rely on equity levels much higher than the low levels current regulations allow, loss absorption would be obtained without any of these disruptions and it would fall most naturally and appropriately by the shareholders who are entitled to the upside. Third, cross-border issues in the resolution of global banks, which played an important role in the Lehman Brothers bankruptcy, have hardly been addressed. If a bank with systemically important operations in different countries goes into a resolution procedure, the procedure will be handled by different authorities in the different countries in which the bank has legally independent subsidiaries; because the different authorities act independently and each authority takes care of problems in its domain, integrated operations in areas such as cash management and IT systems are no longer feasible. It may therefore be impossible to maintain, even temporarily, some of the functions which are essential for the rest of the financial system.67 Flawed Claim 24: If equity requirements are increased, banks will increase their “risk appetite,” which will make the system more dangerous.68 What’s wrong with this claim? As we discuss in Chapter 8, such a claim was made by Bob Diamond when he was CEO of Barclays. Statements like these may be empty threats, but if they                                                                                                                                                                                                 about so-called GLAC “Gone Concern Capital Absorbing Capacity” (e.g., “Progress and Next Steps Towards Ending Too-Big-to-Fail,” Report to G-20, September 2, 2013). 65 See also Simon Johnson, “The Myth of a Perfect Orderly Liquidation Authority for Big Banks,” New York Times Economix, May 16, 2013, and Simon Johnson and Marc Jarsulic, “How a Big Bank Failure Could Unfold, New York Times Economix, May 23, 2013. See also Anat Admati, “Too Much Equity? If Anything, Brown-Vitter Asks Too Little,” American Banker, June 24, 2013. 66 The Bank Recovery and Resolution Directive in the European Union ignores the problem altogether. 67 See Advisory Scientific Committee (2012). The more recent Financial Stability Board’s “Principles for Cross￾border Effectiveness of Resolution Actions,” November 3, 2015includes an enormous wish list and recommendations that would help make cross border resolution viable, but the implementation of these recommendations cannot be expected any time soon. The June 2014 IMF document “Cross‐Border Bank Resolution: Recent Developments” summarizes the key challenges, and they have not been met as of the end of 2015. For the European Reforms, see also Hellwig (2014) 68 See, for example, Bill Black, “Brown-Vitter Will not and Cannot Work but it is Criminogenic,” Naked Capitalism blog, May 11, 2013.

26 are not, the claim raises serious concerns about governance that should trouble banks’ shareholders and boards of directors. If risks are worth taking on behalf of the banks’ investors, why aren’t the banks already taking them? If the risks are not worth taking, why would the banks take them when they are funded with more equity? The claims appear related to the flawed focus on ROE in banking that we discuss in Chapter 8.69 Flawed Claim 25: If equity requirements are increased, bank managers will be less disciplined.70 What’s wrong with this claim? The claim rests on the false notion that bank creditors can “discipline” bankers, or provide better governance, than shareholders, and that bankers are more disciplined when investing borrowed money than when they invest shareholders’ money. The academic literature includes theoretical models that claim to capture the idea that “debt disciplines managers.” Some such theories are specific to banks, arguing that by threatening to withdraw their funding, depositors and short-term creditors can provide “discipline.” As we have argued in various writings, including Chapter 10, these models are a poor basis for policy advice because they lack empirical support and ignore critical elements of the real world which, if included, would reverse their conclusions.71 The fact that assertions about the real world are made on the basis of theoretical models without justifying the appropriateness of the models or addressing the critical issues we raise about their inadequacy is highly disturbing. The suggestion that long-term debt provides better discipline to managers than equity is also flawed in the context of banking. First, whereas long-term debt does not cause a risk of runs, it may still generate systemic risk. As discussed in the context of Claims 21 and 22, if debt holders are sufficiently important for the financial system, for example large insurance companies, it may be deemed undesirable to impose losses on them in resolution or insolvency. Moreover, the too-big-to-fail problem is relevant for long-term debt as well as short-term debt in that the                                                              69 See also Anat Admati, “Beware of Banks’ Flawed Focus on Return on Equity,” New York Times Dealbook, July 25, 2011, and Anat Admati and Martin Hellwig, “The Case Against Banking’s Case for Less Capital,” Bloomberg View, February 5, 2013. 70 A recent example is Raghuram Rajan, “Love the Bank, Hate the Banker,” Project Syndicate, March 27, 2013, which refers to the Washington Mutual (WaMu) bank failure, claiming that it is an illustration that the threat of runs helps provide “discipline” to bank managers. In fact, the timing of the events in the WaMu case is at odds with the argument Rajan seems to be trying to make. Significant withdrawals from WaMu started after the Lehman Brothers bankruptcy on September 15, 2008, and the bank was closed on September 24, 2008. By that time, it was too late to “discipline” the bank’s managers. William Isaac, referenced in Footnote 37 argues that long-term debt provides better discipline than equity. Seemingly echoing such claims, Jamie Dimon, CEO of JP Morgan Chase, warned in 2011 that bankers might do “stupid things” if they had “too much capital.” (See Alistair Barr, “J.P. Morgan’s Dimon concerned about too much capital: Surfeit of capital may make people do ‘stupid things,’ CEO says,” Wall Street Journal MarketWatch, February 15, 2011.) His statement raises the concern of why bankers would do stupid things with shareholder money, and why they would expect to get away with it. 71 We have discussed this problem in earlier writings, particularly Admati et al (2013, Section 5), which first appeared in 2010. In Admati and Hellwig (2013), we explain that fragility in banking is more likely to reflect a lack of discipline, which allows bankers to continue to borrow and thus prevents debt from providing any discipline. See also Admati et al (2015, Section 5), and Pfleiderer (2014).

27 collateral damage associated with distress or insolvency may lead to bailouts. If debt holders believe they can count on being bailed out, they will not impose any discipline on the bank. Second, even if long-term creditors want to impose discipline, the scope for doing so is limited. For example, with a ten-year bond, on average one tenth of the debt is rolled over each year. But discipline can only be imposed when the debt must be renewed and investors negotiate with the bank for the conditions under which a renewal would be granted. As we have argued in the context of the possibility that deposit and short-term debt provide “discipline,” long-term debt may in fact provide the precise opposite of discipline: Negotiating with new short-term creditors, or offering them collateral can make incumbent long-term creditors worse off (should they expect to bear losses), yet these creditors are unable to withdraw their claims until the debt expires. Flawed Claim 26: The best way to make banking safer is to require banks to put funds from deposits into reserves of central bank money or Treasury Bills (so-called narrow banking or the Chicago Plan for 100% reserve banking). Narrow banking will give us a stable financial system, and there would be less need to impose equity requirements. What’s wrong with this claim? Requiring banks to put all funds into cash or Treasury Bills will make these banks safer but the financial system as a whole may become less efficient and/or less safe. If final investors maintain current funding patterns, banks will provide a lot of funding to the government; which may well come at the expense of funding of nonfinancial firms. The experience of southern European countries in the decades before 1990 shows such crowding out of private borrowing by government borrowing can have substantial negative effects on economic growth.72 More likely, narrow banking would lead investors to put substantially more of their money in other institutions, for example money market funds which are “bank-like” without being subjected to the same regulation as banks. As we have seen in the weeks after the Lehman bankruptcy, such institutions can also be subject to runs and can be a major source of systemic risk. Financial instability would merely shift from banks to those “bank-like” institutions. In this context, it is useful to recall that Lehman Brothers was an investment bank, AIG was and is an insurance company and, in Europe, Dexia and Hypo Real Estate were in the covered-bond business; none of the institutions had any deposits. Flawed Claim 27: The financial system would be safe if banks are subject to a 100% reserve requirement so they can take no risk with depositors' money, while non-bank financial institutions are entirely prohibited from borrowing.73                                                              72 See, for example, the essays by Caminal et al., and Borges in Dermine (1990) 73 See Kotlikoff (2010) and Cochrane (2014) for such proposals.

28 What is wrong with this claim? This claim ignores the benefits of using some debt to fund difficult-to-value investments such as loans. Moreover, having no debt in financial intermediation would not necessarily eliminate fragility and possible harm to small investors. Investors want much of their money to earn some interest and yet to be liquid so they can get it fairly reliably when they need it. If banks must operate as open-end mutual funds with no debt, investors who need cash would return (or sell) their shares and get whatever the shares were worth. Determining share values would be easy if the assets held by a fund (of the fund itself) were traded daily on a public exchange, but otherwise would be problematic, and the mutual fund could suffer something similar to runs if shareholders fear significant asset price declines returned their shares and the fund had to sell assets in a hurry.74 Trading in stock markets exposes individuals who need to trade for liquidity reasons to losses from better-informed investors. The opacity of assets consisting of many mortgages and other loans would give rise incentives to those with access to better information to engage in such trading if the shares of banks with 100% equity were traded on stock exchanges. The information-insensitivity of banks' debt is valuable for liquidity provision and the idea of requiring significant equity (such as 30% or even more) but not as much as 100% is intended to preserve this function and strike a balance between liquidity provision and the stability of the banking system. Flawed Claim 28: Tighter regulation of banks, and in particular higher equity requirements, are undesirable because they would cause activities to move to the unregulated shadow banking system.75 What’s wrong with this claim? As we discuss, particularly in Chapter 13, the development of the shadow banking system and the risks it poses point to the past weakness of enforcement. The most dangerous parts of the shadow banking system developed primarily to avoid existing regulation. Examples include the so called off-balance-sheet special purpose vehicles and money market funds, both of which played in infamous role in the 2007-2009 financial crisis. The lessons should be that we need better rules and better enforcement, not that we should give up on rules. Dealing with regulatory arbitrage is challenging, but the challenge can be met, and it must be met if the regulation is important and beneficial.76                                                              74 Gordon and Gandia (2013), for example, show that money market funds with floating value were also quite unstable at the same time that those that promised fixed net asset value were experiencing runs in 2008. Because Germany has had such experiences with open-end mutual funds for real estate investments, the German Federal Ministry of Finance proposed in July 2012 to outlaw open-end mutual funds for real estate investments. 75 See, for example Elliott, referenced in footnote 17. 76 Levitin (2014, p. 2037) asserts that “Admati and Hellwig think that [dealing with the shadow banking system] is easy.” In fact, we have not claimed it is easy to enforce the regulation effectively, only that it is important and possible. In “We are Still Hostages to the Big Banks,” New York Times, August 26, 2013, Anat Admati summarized the response: past failures to make sure that banks could not hide risks using various tricks in opaque markets is hardly reason to give up on essential new regulations. We must face the challenge of drawing up appropriate rules and enforcing them, or pay dearly for failing to do so.”

29 Flawed Claim 29: Since banking is a global business, it is important to maintain a “level playing field”. Therefore, banking regulation must be coordinated and harmonized worldwide.77 What’s wrong with this claim? The claim, as discussed in Chapter 12, is false. If some countries foolishly allow their banks to pursue very risky strategies and to borrow excessively, this is not a reason why other countries should do the same. Each country should be concerned with how much of a risk from its banks it is willing to accept, just as each country has its own building codes, consumer safety standards, environmental regulation and energy policy. We would not allow chemical companies to pollute rivers and lakes simply because the industry maintains that somewhere in the world another country is allowing these things. The search for “level playing fields” in global competition is highly damaging if it leads to a race to the bottom, where each country ends up fighting stricter regulation on behalf of its members of the industry.78 This problem is further discussed in the context of Claim 30. Flawed Claim 30: Stricter national regulation would harm “our” banks; instead we should be supporting them in global competition. What’s wrong with this claim? Like the preceding claim, this claim is false, as discussed in detail in Chapter 12.79 The success of a nation’s banks in global competition is not an appropriate objective for policy. The global economy is not a sports event where a country might win medals in all disciplines. Rather, it is a system in which people and firms from different countries trade with each other, and a country necessarily “loses” in the markets for those goods which it imports. For the country, and for the people living in it, it is efficient to specialize on goods they are good at and to import the others. Government subsidies to banks, or indeed any firms, in international competition is undesirable; such subsidies create distortions in favor of these firms at the expense of others in the economy, and it may direct too many resources, including talent, inefficiently to one industry over others. Weak regulations that allow banks or other firms to take risks at the expense of others are also distorting. It is legitimate for national regulators to protect their citizens by regulating foreign banks’ subsidiaries if regulations in the banks’ home country are deemed to be insufficient or ineffective.                                                              77 This argument is made frequently. See, for example, The Clearing House, referenced in footnote 5, and S&P, referenced in footnote 18. 78 See also Anat Admati and Martin Hellwig, “Global Level Playing Field Arguments are Invalid,” a version of which appeared as a comment in Financial Times, June 3, 2011. (The text is available at http://www.gsb.stanford.edu/news/research/admati-battle-begun html) The Federal Reserve has effectively rejected this notion in other aspects of U.S. financial regulation by mandating the creation of intermediate holding companies to focus all the assets and liabilities of foreign banks operating in the United States to make it harder for these banks to evade national regulation. This model can be extended and applied to other aspects of international banking in a way to reduce the consequences of a failure of international financial regulatory harmony. 79 See also the article referred to in the previous footnote.

30 Flawed Claim 31: The politics of banking makes effective regulation impossible, and therefore debating the merits of specific regulations such as equity requirement is “beside the point.”80 What’s wrong with this claim? This claim, typically made without a suggestion as to how to overcome the political challenge, suggests that there is no choice but to allow flawed claims and dangerous policies to persist. The claim is analogous to saying that “politics makes corruption unavoidable; thus debating the merits of specific anti-corruption strategies is beside the point,” or: “the politics of organized crime makes effective criminal enforcement impossible; thus debating specific strategies for fighting organized crime is beside the point.” Whereas the politics of financial reform (including the outsized influence that banks have on the political process and the symbiotic relations of banks and governments) certainly makes quick progress unlikely, the eventual success of many reform movements has shown that change is possible. Reform, however, requires public awareness and debate, and sensible debate requires understanding of the issues. Clarifying the issues and empowering more people to participate can create public pressure on those who refuse to engage or to take action, and can eventually bring about the necessary political will for better regulation.81 In reviewing our book, Martin Wolf concluded that our views are not more widely accepted because “bankers are so influential and the economics are so widely misunderstood.” His final assessment is that: “we have failed to remove the cause of the crisis. Further such crises will come.”82 Because risk from banking is more abstract than risk from plane crashes or shoddy bridge construction, flawed claims about banking may have more staying power. However, the harm from a distorted and dangerous financial system is large and affects many people. The current regulations can be greatly improved, bringing large benefits to society. And understanding the issues does not require advanced training. If more people understand the issues, we have a chance of getting serious reform.

                                                             80 This claim is made in Levitin (2014, p. 2067), who reviews our book together with others. A few of these books describe the writers’ experiences in politics and regulation. The books by Sheila Bair, Neil Barofsky and Jeff Connaughton, in particular, highlight the political challenge and aim to increase political pressure for reform, but they do not explain the underlying economics in as much detail as we do in our book. 81 We discuss the problem of willful blindness in the preface of the book and of the paperback edition, both of which are available on the book website. 82 See Martin Wolf, “Why Bankers are Intellectually Naked,” Financial Times, March 17, 2013.

31 References

1. Admati, Anat R. (2014a), “The Compelling Case for Stronger and More Effective Leverage Regulation in Banking,” Journal of Legal Studies, 531-561.
2. Admati, Anat R. (2014b), “Examining the GAO Report on Expectations of Government Support for Bank Holding Companies,” Testimony for hearing of Senate Committee on Banking, Housing and Urban Affairs Subcommittee on Financial Institutions and Consumer Protection, July 31.
3. Admati, Anat R. (2016), “The Missed Opportunity and Challenge of Capital Regulation,” National Institute Economic Review, Forthcoming, February.
4. Admati, Anat R., Peter M., DeMarzo, Martin F. Hellwig and Paul Pfleiderer (2013), “Fallacies, Irrelevant Facts, and Myths in the Discussion of Capital Regulation: Why Bank Equity is Not Socially Expensive,” Working paper (first draft August 2010).
5. Admati, Anat R., Peter M., DeMarzo, Martin F. Hellwig and Paul Pfleiderer (2015), “The Leverage Ratchet Effect,” Working paper (first draft March, 2011).
6. Admati, Anat R., and Martin F. Hellwig (2011), “Comments to the UK Independent Commission on Banking,” Submitted July 4.
7. Admati, Anat R., and Martin F. Hellwig (2013), “Does Debt Discipline Bankers? An Academic Myth about Bank Indebtedness,” Working paper.
8. Advisory Scientific Committee (2012), “Forbearance, Resolution, and Deposit Insurance”, Report 01 of the Advisory Scientific Committee of the European Systemic Risk Board, Frankfurt, July 2012.
9. Advisory Scientific Committee (2014), “Is Europe Overbanked?” Report 04 of the Advisory Scientific Committee of the European Systemic Risk Board, Frankfurt, June 2014.
10. Basel Committee on Bank Supervision (2015), “Regulatory Consistency Assessment Programme (RCAP) – Report on risk- weighted assets for counterparty credit risk (CCR),” October.
11. Behn, Markus, Rainer Haselmann, and Vikrant Vig (2014), “The Limits of Model-Based Regulation”, Institute for Monetary and Financial Stability, University of Frankfurt, Working Paper 82/2014.
12. Brandao, Luis Marques, Ricardo Correa, and Horacio Sapriza (2013), “International Evidence on Government Support and Risk Taking in the Banking Sector,” IMF Working Paper WP/13/94.
13. Brooke, Martin, Oliver Bush, Robert Edwards, Jas Ellis, Bill Francis, Rashmi Harimohan, Katharine Neiss and Caspar Siegert (2015), “Measuring the Macroeconomic Costs and Benefits of Higher UK Bank Capital Requirements,” Financial Stability Paper 35, Bank of England.
14. Bulow, Jeremy and Paul Klemperer (2013), “Market-Based Capital Regulation,” Working Paper.

32 15) Calomiris, Charles W. (2013), “Reforming Banks Without Destroying Their Productivity and Value,” Journal of Applied Corporate Finance, Vol. 25 (4), 14-20. 16) Cochrane, John H. (2014), “Towards a Run-Free Financial System,” working paper. 17) Cole, Rebel A. (2013), “How did the Financial Crisis Affect Business Lending in the U.S?” Working paper. 18) DeAngelo, Harry and René M. Stulz (2015), “Liquid-Debt Production, Risk Management, and Bank Capital Structure: Why High Leverage is Optimal for Banks,” Journal of Financial Economics, 216-236. 19) Dermine, Jean (ed.) (1990), European Banking in the 1990s, Blackwell, Oxford. 20) Dowd, Kevin (2015), “No Stress: The Flaws in the Bank of England’s Stress Testing Programme.” Working paper. 21) Elliott, Douglas (2013), “Higher Bank Capital Requirements Would Come at a Price,” Brookings paper, February. 22) FSA (Financial Services Authority) (2010), “The Prudential Regime for Trading Activities: A Fundamental Review, Discussion Paper 10/4, Financial Services Authority, London. 23) FSB (Financial Stability Board) (2015), “Principles on Loss-absorbing and Recapitalisation Capacity of G-SIBs in Resolution: Total Loss-absorbing Capacity (TLAC) Term Sheet,” November. 24) Galbraith, John K. (1954), The Great Crash 1929, Houghton Mifflin, Boston. 25) Gordon, Jeffery N. and Christopher Gandia (2013), “Money Market Funds Run Risk: Will Floating Net Asset Value Fix the Problem?” Working paper. 26) Gorton, Gary, and Andrew Metrick (2010), “Regulating the Shadow Banking System”, Brookings Papers on Economic Activity 41(2), 261–312. 27) Gorton, Gary (2012), Misunderstanding Crises: Why We Don’t See them Coming, Oxford University Press. 28) Gorton, Gary, and Andrew Winton (2014), “Liquidity Provision, Bank Capital, and the Macroeconomy,” Working paper. 29) Haldane, Andrew G. (2012) “The Dog and the Frisbee.” Speech delivered at the Federal Reserve Bank of Kansas City’s 366th economic policy symposium, Jackson Hole, Wyoming, August 31. 30) Haldane, Andrew G. (2013) “Constraining discretion in bank regulation,” speech at the Federal Reserve Bank of Atlanta Conference, April 9. 31) Hellwig, Martin F. (2014), “Yes, Virginia, There Is a European Banking Union! But It May Not Make Your Wishes Come True”, Preprint 2014/12, Max Planck Institute for Research on Collective Goods, Bonn. 32) Hoenig, Thomas (2013), “Basel III: A Well Intentioned Illusion,” Speech to Association of Deposit Insurers, April 9, Basel.

33 33) Jakab, Zoltan and Michael Kumhof (2015), “Banks Are Not Intermediaries of Loanable Funds – and Why this Matters”, Bank of England Working Paper No. 529, May. 34) Kapan, Tümer and Camelia Minoiu (2013), “Balance Sheet Strength and Bank Lending During the Global Financial Crisis,” IMF working paper 13/102. 35) Kerr, Gordon (2011), “The Law of Opposites: Illusionary Profits in the Financial Sector,” Working Paper. 36) King, Mervyn (2013), “A Governor looks back – and forward,” speech given at the Lord Mayor’s Banquet for Bankers and Merchants of the City of London, June 19. 37) Kotlikoff, Laurence J. (2010), Jimmy Stewart Is Dead: Ending the World's Ongoing Financial Plague with Limited Purpose Banking, Wiley. 38) Levitin, Adam J. (2014), “The Politics of Financial Regulation and the Regulation of Financial Politics: A Review Essay,” Harvard Law Review, 1991-2068. 39) Lewis, Michael (2014), “Flash Boys: A Wall Street Revolt, Norton. 40) McLeay, Michael, Amar Radia and Ryland Thomas (2014), “Money Creation in the Modern Economy”, Bank of England Quarterly Bulletin Q1, 14-27, 41) Pagano, Marco, and Roell, Ailsa (1990), “Trading Systems in European Stock Exchanges: Current Performance and Policy Options”, Economic Policy 5(No. 10), April 1990, 63 – 115. 42) Pagano, Marco, and Ailsa Roell (1993), “Front Running and Stock Market Liquidity”, in: Vittorio Conti and Rony Hainaut (eds.), Financial Market Liberalization and the Role of Banks, 178-199. 43) Persaud, Avinash D. (2014), “Why Bail-in Securities are Fool’s Gold,” Peterson Institute for International Economics, Policy Brief PB14-23 44) Pfleiderer, Paul (2015), “On the Relevancy of Modigliani and Miller to Banking: A Parable and Some Observations,” working paper. 45) Pfleiderer, Paul (2014), “Cameleons: The Misuse of Theoretical Models in Finance and Economics,” working paper. 46) Pozsar, Zoltan (2014), “Shadow Banking: The Money View,” Office of Financial Research, Working Paper 14-04, July 2004. 47) Roell, Ailsa (1990), “Dual-Capacity Trading and the Quality of the Market”, Journal of Financial Intermediation 1, 105-124. 48) Tobin, James (1967), “Commercial Banks as Creators of ‘Money’”, in: Financial Markets and Economic Activity, ed. Donald Hester and James Tobin, Cowles Foundation Monograph 21, Yale University Press, New Haven, CT. 49) Vestergaard, Jakob and María Retana (2013) “Behind Smoke and Mirrors: On the Alleged Recapitalization of Europe’s Banks,” DIIS Report.

Dear Committee, I support the proposal that Trading Banks hold higher capital reserves as a buffer against financial shocks. I say this, recognizing the argument raised that this will/may increase interest rates is, I consider real. I am a business owner supplying supermarkets where it is virtually impossible to get a price increase. You may have read that in Australia some suppliers have stopped supplying the supermarkets because they have declined the suppliers price increase. So it is a real issue. Accordingly, increasing the capital ratio will impact our margins if Banks do raise their deposit rates to meet the higher Capital reserves, and consequent lending rates. I would also like to explain I have previously been a Director of two mezzanine lending Finance companies, and thanks to a strong cashflow in our group we were able to progress through the GFC without default. The other factors assisting us were a) Good/adequate lending practices b) Board determined lending policies and LVR ratios which provided a degree of cushioning which extended us to the full extent of the down-turn in value of the security. A further management tool would have been higher capital reserves, as proposed by the Reserve Bank on Trading Banks. In terms of higher capital reserves I would like the Reserve Bank to consider the percentage value accounted to for different classes of asset. Firstly, I am aware that Westpac Bank mis-reported the algorithm they used when reporting their capital ratio to the Reserve Bank. I closed my accounts with them accordingly. This is an error that should have been picked up earlier by the Reserve Bank, and also internally by Westpac – so I favour a system of open disclosure which the Reserve Bank has implemented. I also favour more intense auditing of the banks reporting, and closer review of the algorithms they use. Secondly I would like to propose the Reserve Bank makes ‘significant’ and progressively implements different value percentages for different classes of assets. For example I believe the Banks are only able to value businesses Stock on Hand at 10% of its value, or maybe 20%. While that does seem low, I have no strong argument for or against that – other than to say if a Receiver was appointed and they could only realize 20% value for the SOH value (note: recorded at Cost price, - not Sales Price) then I’d have to say the Receiver would be doing a poor job. There is an argument to say the Reserve Bank could raise the Capital value percentage for Stock on Hand. Of greater impact to the NZ economy would be for the Reserve Bank to lower the % value Banks can accord the value of a Residential house (currently something like 80%?), and increase materially the %value that can be accorded to business lending. I submit, this would have many/multiple benefits:

a) Banks would be keener to lend to business viz a viz residential property as it would account for an increasingly higher % of their capital reporting value. b) This would diminish the banks propensity for residential home lending. c) As this occurred house lending interest rates would rise because it would become harder to get funding, and there would be at least the same demand for a lessor pool of funding available for housing – giving the banks better cashflow from higher interest rate mortgage lending. So what they lose in capital reporting capacity, they gain in cashflow through higher interest rates. d) Higher interest rates will diminish the demand for speculative lending on houses, and assist to drop the inflationary pressure on house price index/house prices. e) Lower demand due to higher interest rates will result in lower demand for housing, making housing more available (both in terms of number of homes available, lower price, but higher debt servicing to first home buyers – who may be able to be given some preferential treatment by the banks/Reserve Bank in terms of lending). This should assist the current debacle of what KiwiBuild is trying to achieve. f) Increasing the Capital % value the banks can apply to business lending will encourage them to reduce interest rates to business. There are real benefits to this i) Banks become better lenders to business. By better I mean they understand business better, and do a better job at business lending – as opposed to our current situation of having ‘many people skilled in the art’ of residential lending. ii) More lending to business will create greater competition amongst the banks, thus causing them to lower business interest rates. (so the banks can secure the higher Reserve Bank capital ratio for this class of lending.) iii) More business lending will create more investment within businesses, growth, greater efficiencies and more employment. iv) More employment will lower State funding of Social welfare and all associated costs. v) Lower interest rates will encourage some ‘current fence sitters’ who are thinking about Private Enterprise viz a viz ‘employment income’ to commit to taking that jump in to creating their own private enterprise business. vi) This should raise NZ’s GDP and create a bigger tax take and a stronger economy. vii) In my mind there should be a reasonable differential between business lending rates and and house lending rates – just as there is now. It should however be in the reverse. For example house price lending is around 4%. Business lending is around 7%. The Reserve Bank should change the capital ratios for these two types of lending, increasing the business lending ratio until house price lending is around 8% at the consumer/home owner level and business lending is no higher than 4%. The nation is then lending on productive assets and not unproductive, inflationary assets such as second homes. viii) Businesses would then be able to issue Bonds, as they currently can at around 6% to arrange additional finance, if/as required. I would be happy to discuss this further if deemed appropriate. Otherwise, I trust you will take this into consideration when reviewing, and I trust implementing higher capital ratios for the Trading Banks.

And please be vigilant in Banks reporting their ratio’s and the use of their algorithms. We simply cannot afford to have any fraudulent activity/reporting in our society whatsoever. Fraudulent Capital reporting, even if deemed accidental, would be far more damaging to those who have entrusted their funds to that bank, than higher or lower interest rates. Yours sincerely Brett N Brett Tawse

OIA s9(2)(a)

Bruce Merrett I am an individual (a small consumer of the banking system) but for me bank stability is very important. . I & need the reassurance that my money is very secure. I willingly will pay a little more in bank fees to secure my funds. I am aware that banks can calculate very accurately the reserves needed for most 'eventualities' but I need the assurance that there is a safer margin than at present & would welcome a compensation scheme similar to the UK where the first $100000 per account say is guaranteed or underwritten by the deposit taker. I am also aware of bigger issues related to FRB etc. but I wanted to put my 'small customer' viewpoint. thanks...Bruce OIA s9(2)(a)

Hi there, my Name is Bryan Patrick Beeston, and my Partner Annette have been borrowers from the banking system since June 1982, with a loan from the rural bank! In this time we have seen many bank mergers, failures, and interest rates from 20+% To today’s low rates. Yet has a business our only concern is the access to funding both short and long. And looking back the only mistake the reserve bank made, was to allow commercial bank to do interest only lending, which in our view allow borrowers to borrow up to 20% more therefore creating inflationary pressure on land, housing, which end up self for filling securities with inflated prices. And since the tightening of borrowing and back to the 20 year pay back of loans has not only dampen the markets it caused a drop in values, and lower the ability to borrow to much, which again hold prices to a realistic value, to what it can earn. So you have already brought the banking business to account, now adding that they carry enough cash to cover a one in 200 year event, is driving the economy to a stand still, has business have to adjust to paying principal back, the banks are looking to there clients to fund the increase capital that the reserve bank is looking to require. And this is making us, and other employers of capital and people choke. Has dairy farmers the reserve bank is looking for a reduction of at least $8 billion. And that is possible in the next 4/6 years. Yet is not help with higher margins that the bank are wanting because of expected new rules. Our banking was fine , and would be in better shape if we not strayed from the 20 year pay back policy. Yours Bryan Beeston. OIA s9(2)(a)

About Buddle Findlay Buddle Findlay is one of New Zealand’s leading law firms with offices in Auckland, Wellington and Christchurch. The firm has a total of 181 lawyers, including 46 partners and, with support staff, a total workforce of 271 people. Our origins date back over 120 years, to the earliest days of legal practice in New Zealand. Buddle Findlay has one of New Zealand’s leading financial services teams. We act for all major New Zealand banks, as well as off-shore banks and other international financial institutions, and for large New Zealand corporates on financing matters. Our team is highly experienced, with some of the partners having worked together for over 20 years. Most of our senior lawyers have worked in global firms, covering all the major financial centres around the world, including London, New York, Hong Kong and Sydney. Our practice covers all aspects of banking and finance including bank capital raising, advice on prudential regulation, bank funding, derivatives, regulatory compliance (Reserve Bank, NZX, Financial Markets Conduct Act, Commerce Act, Fair Trading Act and advertising), payments and clearing systems, securitisation and structured finance, capital markets transactions (debt and equity), as well as corporate and institutional finance and consumer finance.

SCOPE OF SUBMISSION Paper No. 4 specifically seeks views on a range of questions – mostly related to the levels of capital that the Reserve Bank is proposing to require of banks. Our submission, however, only focuses on:

1. Tier 2 capital and why it should be retained as regulatory capital; and
2. although not a separate question, the “in principle decisions” taken in the Capital Review in relation to the form of Tier 1 capital and, particularly, the proposal not to permit contingent convertible debt or redeemable preference shares as Tier 1 capital. While we acknowledge that the questions of what should qualify as bank capital were discussed in Capital Review Paper 2: What should qualify as bank capital? Issues and Options (July 2017) (“Paper No. 2”), we believe that as a result of any decision to substantially increase the capital requirements for New Zealand banks, the “in principle decisions” on what qualifies as capital made following consultation on Paper No. 2 should be reconsidered.1 This is because of the compounding effect of:
3. significantly increasing the regulatory capital requirement for New Zealand banks; and
4. significantly reducing the options for banks to raise that additional capital. Furthermore, we also believe it would be inappropriate to make decisions on how much capital banks should have before the detailed proposals on the specific contractual requirements for the various types of instruments promised in Paper No. 2 have been released and consulted on.2

While extensive economic modelling appears to have been undertaken in relation to the setting of minimum regulatory capital requirements, the risk assessment in relation to the forms of acceptable instruments appears to be done at a theoretical and anecdotal level, with little analysis of the legal framework for the various forms of regulatory capital. 1 Susan Guthrie “The role of preference shares in the capital regime” (Memorandum for FSO, 17 November 2017) (the “Preference Share FSO Paper”). 2 See for example, Paper No. 2 at [184, 193, 198, 202]. 4 Submission

Ultimately, we believe it is the legal framework for the issuing of Tier 1 and Tier 2 capital instruments and how that supports their loss absorbency characteristics that is critical – and, indeed, the most important consideration for determining what should or should not qualify as Tier 1 and Tier 2 capital. This paper sets out why we do not believe that the existing rules relating to Tier 1 and Tier 2 capital need to be changed. Structure of this submission This submission is structured in the following way: Part 1: Executive Summary Sets out our conclusion that there is no compelling reason to change what should be allowed as regulatory capital. Part 2: Legal System Strength Discusses the New Zealand legal framework for bank capital and why it is robust and why many of the concerns raised in Paper No. 2 are unwarranted. Part 3: Tier 1 Capital Specifically comments on the Tier 1 regime and why contingent convertible debt and redeemable preference shares should qualify as alternative Tier 1 capital. Part 4: Tier 2 Capital Sets out why Tier 2 capital is a valuable prudential capital instrument and should be retained. Buddle Findlay 5

PART 1: EXECUTIVE SUMMARY This part sets out a summary of our key submissions as follows: The Reserve Bank must revisit its in principle decision not to allow certain types of alternative Tier 1 capital. The Reserve Bank must revisit its in principle decision not to allow certain types of alternative Tier 1 capital because: a) of the compounding effect of potentially increasing regulatory requirements and reducing access to capital by disqualifying key forms of capital; b) the decision is in part predicated on the Reserve Bank’s light handed approach to regulation, which, since the in principle decision, is now subject to review; and c) the in principle decision should not have been made until after the promised consultation on the detailed terms of the changes to Tier 1 capital rules. Implicit in the decision to disqualify certain internationally accepted alternative Tier 1 instruments is a lack of confidence in New Zealand’s conduct and prudential regulators and the regimes they administer. This is unwarranted. The most important element of prudential capital is that it absorbs losses before depositors and creditors. New Zealand has a robust: a) financial markets conduct regime that effectively eliminates moral hazard arising from mis-selling; and b) insolvency regime (including in the Reserve Bank of New Zealand Act 1989 (the “Reserve Bank Act”)) that ensures legal certainty for the loss absorbing characteristics of all alternative Tier 1 capital. New Zealand regulators have the skills and resources to effectively administer these regimes. There is no reason for New Zealand’s prudential regulator to take a different approach to that accepted internationally by other prudential regulators. 6 Submission

There is no credible reason to regard alternative Tier 1 instruments as not absorbing losses on a going concern basis like ordinary shares. Ordinary shares can be brought back and can potentially be the subject of put options to sister companies to the parent just like redeemable preference shares and so these features should not be used as a reason to disqualify them as Tier 1 capital. In practice, for all forms of Tier 1 capital it is the regulatory restrictions on when distributions can be made and when they can be brought back that is the key determinant of going concern capital. Redeemable shares and contingent capital both have effective regulatory restrictions on distribution and buyback to enable them to be counted as Tier 1 going concern capital. Tier 2 capital should be retained but without the requirement for a non-viability write off mechanism. Tier 2 capital is effective to absorb losses ahead of depositors and hence should be retained as an option at, at least, the current 2% of capital level. It is, we understand, one of the most viable sources of capital, particularly for domestic banks. However, the New Zealand insolvency regime is sufficiently robust that there should be no need for a write off mechanism. Philosophically, Tier 2 subordinated debt should absorb losses after all shareholder funds have been exhausted – not before. Access to capital from diversified sources only strengthens New Zealand’s financial stability. Contingent debt, subordinated debt and redeemable preference shares are a valuable source of capital for banks, giving them options to raise capital from third parties without diluting shareholder interests. We believe the benefit of this is likely to far outweigh any of the anecdotal and hypothetical risks raised for disqualifying them. Buddle Findlay 7

PART 2: LEGAL SYSTEM STRENGTH This part discusses the New Zealand legal framework for bank capital and why it is robust and why many of the concerns raised in Paper No. 2 are unwarranted. THE ROLE OF THE LEGAL SYSTEM Paper No. 4 states that the Capital Review is guided by six high-level principles.3 Based on that list, the key features of regulatory capital from a legal perspective are:4 • Capital must readily absorb losses before losses are imposed on creditors and depositors. • The capital framework should be practical to administer, minimise unnecessary complexity and compliance costs, and take into consideration relationships with foreign-owned banks’ home country regulators. • The capital framework should be transparent to enable effective market discipline Fundamental to achieving these outcomes is the legal system into which the instruments are issued. In particular, if instruments can be issued that “absorb losses before losses are imposed on creditors and depositors”5 with a high level of legal certainty, they are capital. This section of our submission examines the key relevant parts of the New Zealand legal system and sets out why they are sufficiently robust such that there is no need to change the instruments currently permitted as regulatory capital – because in New Zealand they are effective in accordance with their terms. There is no need to make ‘nuanced judgments’ about contingent debt’s ability to absorb losses, because it clearly can. TWIN PEAKS New Zealand operates a “twin peaks” system of financial regulation. This involves the Reserve Bank as the prudential regulator and the Financial Markets Authority (“FMA”) as the conduct regulator. This regime was introduced after the Global Financial Crisis of 2007/2008 (“GFC”) following a report from the Capital Markets Development Task Force.6 It is generally regarded as a world class model of financial regulation and in our view sufficiently robust to enable all forms of commonly accepted 3 Paper No. 4 at [7]. 4 The other three factors go to how capital is calculated at the level of capital required, which we do not comment on. 5 This is the first of the Reserve Bank principles and we believe the most important. 6 Capital Markets Development Taskforce Report (December 2009). 8 Submission

prudential capital to be issued in New Zealand. There needs to be a compelling case for New Zealand not to recognise some forms of capital that are internationally accepted by other prudential regulators. To do otherwise runs the risk of signalling a lack of confidence in the legal system and/ or the regulators overseeing it – neither of which we believe is justified.7 We expand further on this below. CONDUCT The most significant area of change for New Zealand’s financial markets legislation in recent times was a major overhaul of New Zealand’s conduct regulation with the establishment of the FMA (to replace the Securities Commission) and the replacement of the Securities Act 1978 and the Securities Markets Act 1988 with the Financial Markets Conduct Act 2013 (“FMCA”). As a result of these reforms it is generally believed that New Zealand has a world class system of conduct regulation. The recent review of New Zealand financial systems stability by the International Monetary Fund8 made very few recommendations in relation to New Zealand’s conduct regulation (which were largely limited to the reliance on trustees and the lack of a licencing regime for custody services).9 In particular, New Zealand has a very robust and well enforced regime relating to: • the offer, issue and sale of financial products; and • fair dealing rules. It also has a well-resourced and effective conduct regulator in the FMA. Concerns about Tier 1 and Tier 2 capital being “mis-sold” are unwarranted in the New Zealand market and the Italian bank example which was given to support these assertions is not relevant.10 While these concerns may have had some foundation under the old Securities Act regime and when a lesser resourced Securities Commission was administering securities law – these concerns are no longer fair. In our experience, the review process undertaken by the FMA, which occurs either prior to registration of the regulated disclosure (this is at the option of the issuer but is widely utilised) or in any event, prior to the acceptance of any applications (this is legislated in the FMCA as a mandatory “waiting period” while the review takes place) is very robust. The FMA’s focus in that review is twofold; compliance with the strict requirements of the legislation and ensuring that there is nothing in the disclosure documents (including by way of omission) which could mislead or deceive an investor. We believe that the New Zealand market has appropriate inbuilt legal protection, to enable Tier 1 or Tier 2 instruments (including contingent debt) to be effectively sold into both the wholesale and retail market. Any moral hazard associated with those instruments can largely be ignored. To suggest otherwise would amount to an unwarranted lack of confidence in the FMA and in the extensive reforms to financial market conduct after the GFC. 7 See for example, Financial Stability Board Principles on Loss-absorbing and Recapitalisation Capacity of G-SIBs in Resolution Total Loss-absorbing Capacity (TLAC) Term Sheet (9 November 2015) at 11 [6b]. 8 International Monetary Fund New Zealand: Financial Sector Assessment Program: Financial System Stability Assessment (Monetary and Capital Markets Department, Country Report No. 17/110, 8 May 2017). 9 International Monetary Fund New Zealand Financial Stability Assessment (10 April 2017). 10 Paper No. 2 at [33]. Buddle Findlay 9

INSOLVENCY REGIME The second element which is critical to the effect of issuing Tier 1 and Tier 2 capital instruments is the New Zealand insolvency regime. How the instruments perform on insolvency is critical to them acting effectively to absorb losses. New Zealand’s insolvency regime is generally considered to be a robust and effective regime and instruments such as Tier 1 and Tier 2 capital instruments should perform in accordance with their terms on insolvency. Consistent with comparable insolvency regimes in other developed nations, it is a fundamental principle of liquidations under New Zealand law that the application of assets in the insolvent estate proceeds (with limited and well-recognised exceptions) on the basis of a pari passu distribution of assets between unsecured creditors, ahead of the interests of shareholders. Where a creditor has however agreed to accept a lower priority in respect of the claims of other unsecured creditors, under New Zealand insolvency law that subordination will be given effect to in accordance with its terms.11 Furthermore, the Reserve Bank has the option of applying the insolvency regime from the Reserve Bank Act to insolvent banks. This statutory management regime gives the statutory manager wide discretion, for example, to suspend payments.12 This regime also gives the Reserve Bank the power to give a statutory manager a direction13 and so the Reserve Bank is able to ensure that all forms of Tier 1 and Tier 2 regulatory capital perform in accordance with its expectations.14 To the extent the Reserve Bank still has concerns, it has the option of specifically legislating to ensure Tier 1 and Tier 2 instruments have the necessary legal certainty as it has done, for example, with covered bonds. PRUDENTIAL REGULATION The Reserve Bank commented in Paper No. 2 that it has: adopted a principle of being conservative in capital matters in part to counterbalance the relatively light-handed approach it takes in other supervisory areas. This statement pre-dates the 2018 Reserve Bank and Treasury consultation paper15 that examines the role of the Reserve Bank and how it is governed (and implicitly includes the approach which it takes to supervision). It is also clear from that consultation paper that the issue of deposit insurance for New Zealand is also a matter which is going to be consulted on.16 Accordingly, we believe that if the Reserve Bank has concerns that its existing rules in relation to what qualifies as capital (which are based on the Basel standards) are not sufficiently conservative, then, at the very least, it should hold off any decisions on changing those rules until its own role is clarified, and the question of whether deposit insurance is introduced into New Zealand is determined (which could further differentiate deposits from capital in the form of contingent debt). 11 Companies Act 1993, s 313(3). 12 Reserve Bank Act, s 127. 13 Reserve Bank Act, s 120. 14 We, therefore, do not accept it is reasonable for the Reserve Bank to refer to “unexpected legal impediments” as grounds for a more conservative approach to contingent debt as it does in Susan Guthrie’s “Next steps for the capital definition” (Memorandum for FSO, 17 November 2017) at [18]. 15 Reserve Bank and Treasury Consultation: Safeguarding the future of our financial system: The role of the Reserve Bank and how it should be governed (1 November 2018). 16 n 15 at chapter 4. 10 Submission

This is particularly so to the extent that the rationale for not adopting Basel standards is because the Reserve Bank does not regulate in a way which is consistent with other central banks (i.e., it has a much more ‘hands off’ approach).17 If, as a result of the review of the Reserve Bank Act, the Reserve Bank’s approach to prudential supervision changes, then the rationale for adopting more conservative legal standards in relation to the form of Tier 1 and Tier 2 capital instruments may no longer be relevant. While noting the Reserve Bank’s argument about being more conservative,18 there is no objective assessment of the amount of risk that the Reserve Bank believes it is avoiding by adopting a more conservative approach on what it would approve as Tier 1 and Tier 2 capital. Many of the arguments raised are anecdotal and theoretical and, in our view, not relevant in the New Zealand market. Given that all banks are likely to require more capital if the new ratios proposed by the Reserve Bank are adopted, then it is vital to understand the level of risk the Reserve Bank believes is avoided by removing access to certain capital instruments and balance that against the benefits of the New Zealand banking sector having access to a wider range of capital (and the risks if they don’t). CONCLUSION ON LEGAL FRAMEWORK In our view New Zealand has: • a robust market conduct regime that should eliminate moral hazard on Tier 1 and Tier 2 instruments; and • an effective insolvency regime, which, in respect of banks, has extensive powers reserved (indirectly) to the prudential regulator and therefore a high level of legal certainty on how Tier 1 and Tier 2 capital would behave on failure. Furthermore, the Reserve Bank should not make changes to commonly accepted forms of capital in international markets based on its supervisory approach when that approach itself is under review. 17 Paper No. 2 at [81-85]. 18 Paper No. 2 at [81-85]. Buddle Findlay 11

PART 3: TIER 1 CAPITAL This part specifically comments on the Tier 1 regime and why contingent convertible debt and redeemable preference shares should qualify as alternative Tier 1 capital. Paper No. 2 classifies Tier 1 capital broadly into three categories: • common equity; • preference shares; and • contingent debt capital instruments. For the reasons outlined below we believe that all three types of capital should continue to qualify as Tier 1 Capital – albeit with modified ratios of what can be included given the proposal by the Reserve Bank to increase minimum regulatory capital requirements. COMMON EQUITY It is universally accepted that common equity is generally the best form of regulatory capital. However, in our view, the Reserve Bank’s analysis of common equity/ordinary shares proceeds on two fundamental legal misconceptions: • that holders “can only exit their investments by finding someone to sell the share to”;19 • that “the value of undistributed earnings and reserves held as buffers against unexpected loss is less certain than ordinary share capital”.20 While this may have been the case prior to the revision of the Companies Act, neither has been the law in New Zealand since that time. In particular, companies can offer to buy back ordinary shares at any time provided that the company remains solvent after the buy back. The amount paid back to shareholders is not limited to retained earnings and can, for example, include some or all of the price paid by the shareholder for the ordinary shares. Par value as a concept no longer exists in New Zealand law (except to some extent for tax purposes). 19 Paper No. 2 at [9]. 20 Paper No. 2 at [13]. 12 Submission

21 n 1 at [6]. In respect of the ability to be repurchased, common equity or ordinary shares are not materially different from contingent debt and/or preference shares which are solely callable at the option of the company. The perceived permanence of common equity when compared to these other types of instrument is a fiction. Companies can buy back all of these types of instruments from holders. The only difference between contingent debt and/or preference shares and common equity is that companies can generally compel buy backs in respect of the former, but can only offer to buy back common equity if the company is solvent. Furthermore, the “redemption in name” concern raised by the Reserve Bank in respect of non￾redeemable preference shares can just as easily be applied to common equity – shareholders could be given a put option at an agreed price or date to sell ordinary shares to a sister company of a bank shareholder.21 Accordingly, “redemption in name” is not a credible reason for differentiating non￾redeemable preference shares from ordinary shares as capital. The restriction imposed on buy back in a bank’s conditions of registration on when capital can be redeemed is, in practice, the fundamental restraint. It does not differ (in substance) between different types of Tier 1 capital. For example, for practical purposes a bank’s ability to buy back ordinary shares is limited to circumstances when it would still have the minimum equity required in its conditions of registration after the buyback (not compliance with the solvency test). When cash is first paid for shares its value as capital may be more certain. However, that quickly changes as cash is converted to assets and those assets may have variable value. In a legal sense, capital is simply the net assets of the company. While in an accounting sense shareholder funds are broken down to (amongst other things) retained earnings and paid in capital, in a legal sense shareholders entitlement is simply to the net assets. For prudential purposes, there should also be no difference in substance between ordinary shares subscribed for and retained earnings. ALTERNATIVE TIER 1 CAPITAL In our view, once it is accepted that companies/banks can effectively “call” ordinary shares issued, the moral hazard risk with redemption of preference shares and contingent debt capital instruments become less relevant. The ability to be redeemed or repaid is not a differentiating characteristic (in a legal sense) of whether it has debt or equity characteristics. We believe that it is critical to consider the outcomes that are required of prudential capital and then review the proposed instruments in light of those outcomes. As discussed, in our view these are the key requirements identified by the Reserve Bank being: • capital must readily absorb losses before losses are imposed on creditors or depositors. • the capital framework should be practical to administer to minimise unnecessary complexity and compliance costs, and take into consideration relationships with foreign-owned banks’ home country regulators. • the capital framework should be transparent to enable effective market discipline. Buddle Findlay 13

22 Paper No. 2 at [188]. 23 Paper No. 2 at [62]. LOSS ABSORPTION For the reasons outlined in Part 2 of this submission (legal system strength), we believe that it is possible to issue alternative Tier 1 instruments and be confident that they will absorb loss on a going concern basis with a high level of legal certainty. This includes redeemable preference shares and contingent debt instruments. In this respect we do not agree with the Reserve Bank that the “loss absorbing quality of ordinary shares is far greater than that provided by contingent debt”. 22 This is because: • it is entirely within the control of the Reserve Bank as regulator to ensure that contingent debt does provide capital to a failing bank at the requisite trigger points (for example if it is written-off, it effectively becomes retained earnings and clearly Tier 1 capital – because retained earnings are Tier 1 capital); and • as a matter of law, as long as the contingent debt cannot be repaid after a certain trigger point has been reached, it is effectively acting as capital – the right to make repayment is suspended and potential income distributions on those instruments will also be suspended (indeed, the same regulatory restrictions on payment of dividends or redemption must be imposed on ordinary shares after a trigger event otherwise legally dividends can be paid and they can be redeemed as long as the bank is solvent afterwards). While there is often an expectation that contingent debt or redeemable preference shares will be repaid at some point (usually sometime after five years), there is also a clear understanding and legal requirement that this can only occur if the bank is in excess of its minimum capital ratios at the time. COMPLEXITY AND TRIGGER POINTS It is the legal terms of the instrument that enable it to act as going concern capital, particularly in the case of contingent debt. The key term is the point at which the write-off is triggered (effectively converting the instrument into retained earnings). The triggers and terms do not need to be complex or difficult to administer. Many of the concerns raised by the Reserve Bank relate to how the trigger mechanisms work and the impacts on the financial health of a bank where the trigger is activated. As a matter of practice, any resolution applied to a bank runs the risk of exacerbating that bank’s problems. This is best dealt with by having well thought out resolution and recovery policies which give the central bank a wide range of tools to deal with a financially distressed bank, some of which may be more public than others. In our opinion, it is important that the central bank has this in place before changing the rules on what may or may not qualify as Tier 1 or Tier 2 capital, including the steps it will take as trigger points are reached. However, any prudential regime that relies heavily on market discipline and hence transparency (as New Zealand does) runs the risk that distress will be signalled to the market. Breaches of regulatory capital ratios, regardless of whether they also create triggers for contingent debt, will signal a bank distress. While the Reserve Bank specifically expresses some concern that the suspension of payments on capital instruments could signal distress to the market, the example it gives is Deutschebank.23 In effect, the concern is that not paying coupons led to a sell-off in Deutschebank’s shares. However, in New Zealand only one bank is listed – Heartland Bank. Given the relatively small size of Heartland Bank the risk of a share sell-off forcing change in regulatory rules relating to capital coupon payments in New Zealand seems remote. 14 Submission

In conclusion, we believe that the existing regime for alternative Tier 1 capital is robust. The legal (and moral hazard or economic) risks associated with instruments that the Reserve Bank is considering now disqualifying is, in our opinion, very low. We believe that this risk is likely to be outweighed by the benefit of allowing banks access to more diversified capital options. It is vital to enable banks to continue to raise capital using instruments which are currently accepted in the market (domestically and internationally) – typically being instruments that have a redemption option (subject to regulator approval).24 There is no legal reason why such instruments do not meet the principles the Reserve Bank uses to determine regulatory capital and why they should not be effective as alternative Tier 1 capital as is the case in most other similar jurisdictions to New Zealand. MARKET DISCIPLINE We believe that enabling banks to issue Tier 1 and Tier 2 capital instruments (which meet loss absorbency requirements consistent with international principles) will only enhance market discipline. This is particularly so in respect of wholesale issues of Tier 1 and Tier 2 capital instruments. The holders of those instruments will be acutely conscious of any risk taking undertaken by the relevant bank, given that they will bear losses, potentially even in advance of ordinary shareholders. This creates incentives for them to monitor management and the financial position of the banks in which they invest. While they may have limited tools to effect change (in a legal sense) – in a practical sense, banks access to ongoing capital is likely to be significantly constrained if wholesale holders of Tier 1 or Tier 2 capital have concerns about the financial performance of the bank. This should provide incentive enough. To the extent that these instruments are listed they may potentially create incentives for fund managers and share brokers to undertake research on the value of the instrument. While not as transparent an indicator as share price they could nevertheless provide some transparency on the financial position of a bank. Ensuring that banks can issue instruments that are attractive to the market but have the necessary loss absorbency features would seem only positive from a market discipline perspective. THRESHOLD FOR ALLOWING ALTERNATIVE TIER 1 CAPITAL In addition, given the current state of capitalisation of most New Zealand banks, the point at which banks can access alternative Tier 1 capital could be significantly higher than internationally accepted rules. For example, the Reserve Bank could require domestic systemically important banks to hold: • say, 12% of their capital as common equity; and • 4 - 6% of their capital as alternative Tier 1 capital with the option of including up to 2% (the current amount) of Tier 2 or gone concern capital which is still effective in absorbing losses at the point of failure. In short, to the extent there is risk with alternative Tier 1 capital or Tier 2 capital (and we believe the risk is very low, if it exists at all) this can be mitigated by the high levels of common equity required relative to international norms. It is just not appropriate to require all capital to be common equity and deny banks access to diversified capital markets; that should only act to strengthen the resilience of the New Zealand financial system. 24 Reserve Bank of New Zealand (Prudential Supervision Department) Document BS2A: Capital Adequacy Framework (Standardised Approach) (November 2015) at subpart 2B, cl 10b(1)(c). Buddle Findlay 15

PART 4: TIER 2 CAPITAL This part sets out why Tier 2 capital is a valuable prudential capital instrument and should be retained. We believe that it is important that all banks continue to have access to Tier 2 capital – but particularly New Zealand owned banks that have greater constraints on access to capital. The Tier 2 capital market is generally well understood by New Zealand retail investors and so is particularly vital to those banks that don’t have the same access to international capital markets as others. There is a well-established market for it in New Zealand. Accordingly, we believe the risks that it does not behave exactly in accordance with its legal terms is remote. As we noted earlier in this submission, New Zealand has a robust market conduct regime and an effective insolvency regime and for that reason there should be no concern that Tier 2 capital instruments will not operate as expected and, in particular, will not be available to absorb losses. Even though Tier 2 capital only operates as “going concern” capital because it is only written-off at the point of non-viability, it is still effective in absorbing losses before creditors and depositors. This is because Tier 2 capital is clearly subordinated to creditors and depositors under its terms. Tier 2 capital, therefore, meets the key test of regulatory capital. To the extent that the Reserve Bank is concerned that Tier 2 capital is written-off before ordinary share capital (and that this is a potentially perverse outcome), the Reserve Bank could consider adding a requirement to Tier 2 capital to the effect that all rights to distributions and payment are suspended at the point of non-viability. Rather than write-off the amount, the holders of Tier 2 debt could subsequently prove in liquidation if the bank is liquidated and recover amounts in priority to ordinary shareholders. We would support that outcome as it is more consistent with the legal principles relating to insolvency where ordinary shares are lowest ranking. This outcome would also mean the incentives to mitigate loss in the bank are clearly with the shareholder, who has the most capacity through the bank’s board to manage that risk. The outcome also avoids the perverse situation where, as a bank approaches non-viability, shareholders can take risks knowing it is not their capital at risk, but rather the capital of the Tier 2 holders. 16 Submission

25 n 1, at [45]. We also note that Basel III standards allow instruments that do not include a non-viability trigger to be Tier 2 capital if legislation imposes losses on subordinated debt when a bank becomes non-viable. This point is recognised in the Preference Share FSO Paper.25 As indicated in our discussion on the strength of New Zealand’s insolvency regimes earlier in this submission, we believe that sufficient legislation exists in New Zealand. Accordingly, in our view subordinated debt without a non-viability trigger should be allowed as Tier 2 capital in New Zealand. Buddle Findlay 17

AUCKLAND PwC Tower, 188 Quay Street PO Box 1433, Auckland 1140, New Zealand P. 64 9 358 2555 WELLINGTON Aon Centre, 1 Willis Street PO Box 2694, Wellington 6140, New Zealand P. 64 4 499 4242 CHRISTCHURCH 83 Victoria Street PO Box 322, Christchurch 8140, New Zealand P. 64 3 379 1747

17 May 2019 Ian Woolford Financial System Policy and Analysis Department Reserve Bank of New Zealand PO Box 2498 Wellington 6140 via email: CapitalReview@rbnz.govt.nz BANK CAPITAL REVIEW Introduction

1. Thank you for the opportunity to submit regarding the Reserve Bank of New Zealand’s ‘Capital Review Paper 4: How much capital is enough?’

2. The Wellington Chamber of Commerce (the Chamber) has been the voice of business in the Wellington region for 160 years since 1856 and advocates for policies that reflect the interest of Wellington’s business community, in both the city and region, and the development of the Wellington economy as a whole. The Chamber is accredited through the New Zealand Chamber of Commerce network.

3. Business Central represents business interests throughout central New Zealand from Taranaki across to Gisborne and down to Nelson. Business Central is one of the four regional organisations comprising New Zealand’s peak business advocacy group, BusinessNZ. In Wellington, our organisation operates the Wellington Chamber of Commerce, accredited to the New Zealand Chamber of Commerce network. Our organisation also delivers ExportNZ to Wellington and the Hawke’s Bay.

4. As a member of Business NZ, we support their submission. We particularly support their view that increasing banks’ cost of capital will result in higher borrowing costs to our members. In addition, we have included below additional feedback particularly important to our members.

5. The fundamental trade-off being considered is risk versus cost. How much risk is New Zealand prepared to carry of a bank failure, and how much cost is the country prepared to shoulder to mitigate this risk? Relatively recent domestic and international experience demonstrates this is not a hypothetical risk. However, while the costs of a bank failure were clearly illustrated by the South Canterbury Finance experience, among others, it must be acknowledged that mitigating this risk is also very costly.

6. The best way to analyse trade-offs is to conduct a cost-benefit study and publish the results for stakeholders to consider. Unfortunately, such analysis of higher bank capital requirements is lacking to-date in the Reserve Bank’s document.

7. This lack of analysis leads to the question of what has been the Treasury’s role in prudential policy setting to date? It is understandable that the Reserve Bank does not have this expertise in-house. This makes it even more important for the Treasury to be able to analyse the policy proposals, such as the Treasury does for many other government policy proposals with an economy-wide impact. We recommend the Treasury is invited to conduct a thorough cost-benefit analysis as soon as possible and the Bank Capital Review process is paused while this takes place.

8. It is essential that any such Treasury analysis includes the full suite of proposed policies affecting prudential regulatory settings, so as to provide a thorough and co-ordinated assessment of the Reserve Bank’s package. For example, it should include assessment of a deposit insurance regime. Both higher bank capital requirements and deposit insurance will increase the cost of capital in New Zealand, the combined effect will have a material impact on the economy making such analysis vital.

9. An increase in families’ and households’ mortgage servicing costs will have a direct impact on their disposable income and, therefore, their wellbeing. For businesses, they will face a higher cost of capital, resulting in lower returns and slower economic growth. It will flow into reduced hiring intentions or delaying expansion into international markets. Many SME owners finance their working capital through personal home loans, so increased financing costs directly affects their bottom line. It could lead to banks deciding not to lend at all to some customers in this high-risk market segment.

10. These affects will cascade into other policy areas too, such as housing and social mobility. For example, if the costs of borrowing are higher than they otherwise would be, it follows that home-ownership will be lower. This runs counter to competing government priorities such as boosting housing affordability and encouraging home ownership. This reinforces the need for a wider analysis of the policy’s effectiveness and benefits, an analysis which is best undertaken by the Treasury given their much wider policy mandate discussed above.

11. Higher borrowing costs are not theoretical. For example, UBS calculates the Reserve Bank proposals could add between 80 and 125 basis points to mortgage costs because of higher capital costs to the banks. UBS estimates the proposals will result in New Zealand having the highest bank capital requirements in the world, overtaking Norway. This is a financial cost that will be borne by the New Zealand economy year after year. Does New Zealand’s banking industry risk profile justify the need for such stringent capital requirements? How much risk mitigation is too much?

12. Given the points raised in our submission, Business Central supports BusinessNZ’s view not to support any increase in bank capital requirements. Yours sincerely,

John Milford Chief Executive Wellington Chamber of Commerce, Business Central

Submission by to the Reserve Bank on the Capital Review Paper 4: How much capital is enough? May 2019

2 CAPITAL REVIEW PAPER 4: HOW MUCH CAPITAL IS ENOUGH? SUBMISSION BY BUSINESSNZ1 1.0 INTRODUCTION 1.1 BusinessNZ welcomes the opportunity to comment on the Capital Review Paper 4: How much capital is enough? (“the Review Paper”). 1.2 BusinessNZ considers sound macro-prudential policy is important to the entire economy, with minimising risks to the banking system fundamental to the soundness of NZ’s financial system. Source: Reserve Bank of NZ 1.3 The NZ financial system came through the Global Financial Crisis (GFC) in reasonably good shape compared with many other countries, in no small part reflecting the quality of NZ’s regulatory systems. Certainly, there was fall-out associated with the collapse of several finance companies but overall, the financial system managed reasonably well. 1.4 However a reasonable performance notwithstanding, it is entirely appropriate for the Reserve Bank to look seriously at the soundness of its current prudential management systems to see if anything more can be done to manage risk successfully. But in doing so it should keep in mind the optimal amount of resource available for reducing risk, given risk cannot be completely eliminated or if can, not without great cost. For example, imposing greater capital requirements on banks could boost financial stability but might well have the significant disadvantage of increasing the aggregate cost of capital. 1 Background information on BusinessNZ is attached as Appendix 1.

3 1.5 The rationale for the changes is summarized on p.5 of the Review Paper: “The Reserve Bank is proposing this change to reduce the chances of banks failing in New Zealand. If banks in New Zealand fail, some of us might lose money and some of us might lose jobs. However, there would also be indirect costs on all society that may be harder to see that would negatively impact the well-being of all New Zealanders. In the end, we would bear the cost of bank failures, in one way or another.” 1.6 Given the Reserve Bank’s recent discussion document on other significant issues, e.g. the proposal for mandatory depositors’ insurance, care must be taken that in total, such actions do not lead to regulatory overkill. That could simply add to the cost of capital and/or reduce returns to depositors while potentially reducing the credit available both to smaller business and to particular sectors of the economy (outlined below). 1.7 From an economic perspective, risk involves two considerations:

1. The need for more resources, including time and money in order to reduce the amount of risk; and
2. How people will react – likely to indicate an acceptable level of risk well short of zero - in light of what must be given up in order to meet the increased cost of whatever is seen as desirable 1.8 BusinessNZ would welcome the opportunity to discuss the content of this submission, if the Reserve Bank felt this would be useful. RECOMMENDATIONS BusinessNZ recommends that: Any moves towards the adoption of increased bank capital requirements should be made with caution, given the likely unintended consequences outlined in this submission. As a minimum, the Reserve Bank should undertake a comprehensive cost/benefit analysis of the proposals before any further steps are taken. BusinessNZ recommends that: Capital requirements are monitored and developed to ensure consistency with international capital standards, and that the analysis for this is provided by independent advisors. Given much of SME business capital funding is sourced through the banking sector we would not like to see NZ business borrowers having to pay unnecessary additional capital or borrowing costs where this is not commensurate with risk.

4 2.0 GENERAL DISCUSSION 2.1 Before coming to any decision as to the merits or otherwise of increasing bank capital requirements, it is crucial policymakers take a step back and ask some fundamental questions. These include – but are not limited to: • Is there a problem in New Zealand with current capital requirements; (i.e. is there evidence of “market failure” which needs to be addressed)? • If there is a problem, is the problem significant? • What are the costs and benefits (including unintended costs) of increasing capital requirements? • Will the proposal to adopt greater capital requirements effectively address the alleged problem of financial risk (and if so at what cost)? • Are there potential options for improving outcomes which don’t impose significant costs (e.g. by improving information to market participants)? 2.2 In order to justify government intervention, there must be a clear case of market failure and the market failure problem must be significant. Moreover, there is a need to be certain any regulatory action taken will address the alleged problem in a cost￾effective manner. 2.3 Given markets are generally faster at self-correcting than are government intervention efforts, the onus must be on government to prove beyond reasonable doubt that the benefits of greater intervention (increased capital requirements) will exceed the cost, including any unintended cost, consequent upon regulation. 2.4 Regulators usually have strong incentives to minimise their own risk by imposing higher standards than might otherwise be justified. Because regulators do not bear the costs of their decisions (costs will ultimately be passed on to investors/consumers), they may over-regulate rather than take account of the cost/quality trade-offs investors/consumers are willing to make. Given each is unique, individuals will have different risk profiles - some will pay a considerable amount to minimise risk, others will want to invest little in reducing either real or perceived risk. 2.5 The Review Paper proposes the doubling of banks’ minimum “high quality” capital requirements in order to reduce the risk of bank failure to once every 200 years. As most banks at present carry significantly more capital than the minimum level government currently requires, a doubling of existing capital held will not be needed. The Review Paper suggests overall capital requirements would have to increase between 20 percent and 60 percent; this would represent about 70 percent of the banking sector’s expected profits over a five-year transition period.

5 2.6 Arguably, increasing capital requirements will potentially reduce the risk of bank failure and therefore the potential political pressure on the government (the taxpayer) of the day to fund bail-outs (however exceptional the circumstances might be). Nevertheless, an increase would come at a cost, both in respect to economic growth (as outlined in the Review Paper) and the potential impact of restrictions on lending (outlined below). The issue of moral hazard also has potential here – with less risk of bank failure reducing incentives on individuals and investors to monitor bank behaviour. 2.7 While the Review Paper indicates the proposal to increase capital requirements significantly is targeted at minimising the risk of bank failure to a 1 in 200-year event, nowhere in paper is there any detailed evaluation of the proposal’s costs and benefits. Why 1-in-200 years? Why not 1-in-1000, or 1-in-10? 2.8 The Review Paper states that a 0.5 percent annual risk tolerance is embedded in insurance solvency standards across Europe but this has little, if any, relevance to NZ conditions. 2.9 It is acknowledged that the (overseas) studies cited suggest a strong relationship between capital requirements and lowering the risk of bank failure but as the Review Paper states in para 50 – p.22, “…..it is important to also consider the New Zealand￾specific context.” 2.10 Over the last 3 decades, New Zealand has been held up as having developed world class fiscal and monetary policy settings. These include the Reserve Bank Act’s independent monetary policy and the country’s fiscal policy, as espoused by the Fiscal Responsibility Act (now superseded by the Public Finance Act). Successive governments have, over the years, largely endorsed both these essential pillars of NZ’s economic policy, with only minor tweaks here and there. 2.11 As mentioned earlier, the NZ financial system came through the Global Financial Crisis in reasonably good shape compared with many other countries, in no small part reflecting the soundness and quality of its regulatory systems. Certainly, there was fall-out associated with the collapse of several finance companies but overall, the financial system managed reasonably well. Areas of actual weakness have been largely dealt with by imposing greater transparency and reporting requirements on finance companies. 2.12 The Swiss investment bank UBS has published a research paper outlining the possible impact on consumers of what is proposed, particularly the increased cost of capital. According to UBS, the proposals could add between 80 and 125 basis points to mortgage costs as banks attempt to claw back the added capital costs.

6 2.13 Moreover, the UBS study estimates the proposals would see NZ overtake Norway in having the highest bank capital requirements in the developed world. 2.14 The Reserve Bank Governor has acknowledged increased capital requirements would add around 20 to 40 basis points additional premium between the costs at which banks borrow and lend. The increase would be passed on to borrowers as banks would have to raise equity, not just leverage debt. On the other hand, the Reserve Bank considers the increased capital requirement would reduce the risk of bank failure while lower risk would make it cheaper to attract equity since investors’ expected rate of return would be lower as their institution would be safer. While this is possible, it still does not reduce the potential for simply passing the increase on to borrowers and small businesses as an added cost of borrowing or potentially, as a restriction on the ability to borrow through credit rationing (outlined below). 2.15 While the Reserve Bank has been keen to promote the argument that borrowing costs might actually be lower and returns on capital might not need to be so high if investors have confidence their investment is very secure, it is noteworthy that at least 2 credible international credit rating agencies have questioned whether in fact this would be the case if capital requirements were increased. 2.16 Standard & Poors has calculated that increased capital requirements by themselves will not impact on the four “big” banks’ credit rating. And if their credit ratings are not lifted, it appears unlikely they will get the cheaper credit the Reserve Bank appears to suggest. 2.17 In December 2018, Fitch Ratings said the proposals were “radical” and “highly conservative relative to international peers” but that the result would ultimately be “significantly stronger buffers” against system shocks. 2.18 The potential impact on capital markets of raising the required capital could be significant when $15-$20 billion is likely involved. This amount of capital raising in not insignificant and as in reality New Zealand is not a capital rich country, the additional capital would likely have to come from offshore. Given economies of scale and the smallness of the NZ economy by international standards, it is also likely a risk premium would be involved. And while it is accepted the Reserve Bank proposes a phased-in approach over 5 years, the potential impact on capital raising should not be underestimated. To put this in context, the total capitalisation of the NZ sharemarket is only around $150 billion. 2.19 The proposed increase could potentially affect lending in two important ways. First, banks could lend less, thereby reducing the need for capital enhancement. Second, they could try and pass on some of the additional cost to borrowers making mortgage finance both increasingly difficult and expensive to obtain.

7 2.20 It is important that any changes to macro-prudential policy tools the Reserve Bank might make should reflect the above points, particularly where a proposed change could have an unintended impact, including on economic efficiency or equity. 2.21 While it almost goes without saying that the “benefits of regulation must outweigh the costs” if regulation is to be justified, it is also important to determine not only total costs and benefits (including potential unintended costs and/or benefits) but also where expected costs and benefits might fall. For example, in the case of higher bank capital requirements, the benefits (if any) might be widely dispersed but the costs fall on particular sectors. The distributional and equity effects of such a proposal on particular groups must be considered. 2.22 The distributional and equity effects would likely have an added impact on the NZ economy given households’ relatively high net debt and particularly agricultural sector debt. Household debt levels are currently in excess 160 percent of disposable income (compared to around 100 percent in 2000 – see graph below). 2.23 Meanwhile. agricultural debt has continued to ratchet up and is currently sitting at around $63 billion (up from $12 billion in 2000), with around two-thirds of debt focused on the dairy sector. 2.24 Any significant change in interest rates could be enough to tip households and businesses over the edge with flow-on effects for creditors and the economy in general through reduced levels of activity.

8 2.25 While agricultural debt has increased five-fold over the period, other sectors have also shown significant increases but not to the same extent, with household debt having increased nearly four-fold and business debt increasing nearly three-fold over the same period (see table below) Sector Lending (registered banks and non-bank lending institutions) Source: Reserve Banks of NZ 2.26 A 100 basis point increase in interest payments of household debt would add about $2.6 billion on annual interest rates cost facing households while a 50 basis points increase would add about $1.3 billion to interest costs. In respect to the agricultural sector, a 100 basis point increase in interest rates would add around $630 million in interest costs facing the agricultural sector while a 100 basis point increase in business debt would add in excess of $1 billion to interest costs. 2.27 Given the significance of the agricultural sector to the NZ economy (unique in the world compared with most developed countries), it is important restrictions on lending to that sector and/or increasing the cost of capital do not put added pressure on specific sectors, the more so in light of international trading arrangements’ current uncertainty.

9 2.28 Home-ownership, generally speaking, has a number of benefits so that restricting the ability of certain groups to enter the market without adequate cause is problematic. It also acts against the current and previous government’s promotion of mechanisms to increase the home ownership rates of, particularly, younger lower income people. As well, the impact on the availability and cost of small business finance could similarly be problematic. 2.29 Many small business owners use housing mortgage finance to partially fund business activities. They may do this for a variety of reasons including, but not limited to, the fact that housing mortgage finance is generally less costly than business finance. By in effect restricting this source of finance, the ability of many small business ventures to get off the ground could be unnecessarily restricted. 2.30 According to the Ministry of Business, Innovation and Employment, 97 percent of New Zealand businesses employ fewer than 20 people. Many small businesses have little in the way of assets to offer as collateral for loans, so often personal housing is used as security. 2.31 The UBS study states that: “In effect, we estimate New Zealanders could end up paying between NZ$1.9 billion and $2.7 billion more on their home loans each year to fund the additional capital requirements proposed by the RBNZ to reduce the theoretical probability of a financial crisis from once-every 100 years to once-every￾200 years.” (p.8) 2.32 While there would likely be some political pressure on banks not to pass on added costs, either through direct regulatory constraints or other political means, there is a risk of mechanisms to try and restrict consumer cost increases affecting consumers in other ways. Greater bank restrictions on loans to potentially marginal customers would be the most obvious. 2.33 It is possible the Reserve Bank could lower the Official Cash Rate (OCR) in an attempt to compensate for any impact a mandatory capital requirement increase might have on mortgage interest rates but this would not necessarily affect restrictions on lending requirements for specific sectors. It would also call into question the rationale for greater capital requirements if the immediate effect were to increase mortgage interest rates. Such matters would need to be considered as part of the cost-benefit analysis BusinessNZ recommends the Reserve Bank undertake before deciding to substantially raise banks’ minimum capital requirements. 2.34 Further, with OCR currently sitting at 1.75 percent, there is little real room for any significant reduction as the Reserve Bank should recognise the need to leave a little in the tank (i.e. ability to reduce the OCR) should the NZ (or world economy), face a significant downturn. Reducing the OCR to compensate for placing greater capital

10 requirements on banks would be like selling the family silver for a substantially discounted price. 2.35 That the proposed capital requirements are directed only to banks, with other financial institutions arguably not affected, could be considered inequitable and encourage greater lending in less regulated sectors of the economy - not necessarily desirable. However, it is accepted that as NZ’s four “big” banks provide most of the country’s mortgage finance, the failure of smaller lending institutions would be unlikely to cause the NZ economy much financial instability. More likely, the net effect would be an increase in the cost of capital for sectors considered to pose a risk to the financial system. 2.36 BusinessNZ strongly adheres to the idea of travelling up the regulatory pyramid, that is, considering non-regulatory options first, moving “up the pyramid” to generic light￾handed options and introducing more stringent measures only if clearly warranted. 2.37 Since different banks will have different market shares and often specialise in lending to specific sectors, the implications for the banks themselves will need to be examined carefully to avoid any undue impact both on particular banks and on the sectors they lend to. 2.38 Arguably, a reasonable amount of time will be needed if undue disruption to banks, individuals and businesses is to be avoided, allowing for planning ahead with some certainty knowing the rules are not going to change abruptly. 2.39 BusinessNZ accepts that at this stage of the policy development process, a proper understanding of the proposal’s costs and/or benefits might be lacking but considers it crucial to obtain reasonably accurate information before proceeding further. The process should not continue on the basis of a “perceived” problem without first establishing both actual and to the extent possible, unintended costs. 2.40 Given the points raised above, BusinessNZ does not support any increase in bank capital requirements. This conclusion notwithstanding, Business NZ considers there could be merit in providing individuals and businesses with greater information about the risks of investing, potential returns and the need to ensure the use of adequate risk management techniques to minimise the possibility of significant failure. The information campaign should not be directed only to bank depositor risk but also to risk generally, particularly as it appears many are not fully informed of the benefits and costs associated with property or health insurance or the many other risks individuals and businesses face daily.

11 BusinessNZ recommends that: Any moves towards the adoption of increased bank capital requirements should be made with caution, given the likely unintended consequences outlined in this submission. At minimum, the Reserve Bank should undertake a comprehensive cost/benefit analysis of the proposals before any further steps are taken. BusinessNZ recommends that: Capital requirements are monitored and developed to ensure consistency with international capital standards, and that the analysis for this is provided by independent advisors. Given much is SME business capital funding is sourced through the banking sector we would not like to see NZ business borrowers having to pay unnecessary additional capital or borrowing costs where this is not commensurate with risk.

12 Appendix One - Background information on BusinessNZ BusinessNZ is New Zealand’s largest business advocacy body, representing: • Regional business groups EMA, Business Central, Canterbury Employers’ Chamber of Commerce, and Employers Otago Southland • Major Companies Group of New Zealand’s largest businesses • Gold Group of medium sized businesses • Affiliated Industries Group of national industry associations • ExportNZ representing New Zealand exporting enterprises • ManufacturingNZ representing New Zealand manufacturing enterprises • Sustainable Business Council of enterprises leading sustainable business practice • BusinessNZ Energy Council of enterprises leading sustainable energy production and use • Buy NZ Made representing producers, retailers and consumers of New Zealand-made goods BusinessNZ is able to tap into the views of over 76,000 employers and businesses, ranging from the smallest to the largest and reflecting the make-up of the New Zealand economy. In addition to advocacy and services for enterprise, BusinessNZ contributes to Government, tripartite working parties and international bodies including the International Labour Organisation (ILO), the International Organisation of Employers (IOE) and the Business and Industry Advisory Council (BIAC) to the Organisation for Economic Cooperation and Development (OECD).

Callum Streeter I am a New Zealander living in the US and I strongly feel New Zealand needs a Government Sponsored Deposit Guarantee like the FDIC insurance in the US. The level of comfort it gives is great and creates accountability from the Government to ensure financial stability. Thanks, Callum OIA s9(2)(a)

To: Reserve Bank of New Zealand On: Capital Review Paper 4: How much capital is enough? 17 May 2019

RESERVE BANK: CAPITAL REVIEW PAPER 4: HOW MUCH CAPITAL IS ENOUGH? | 3 However we sense, from the tone of the Consultation Paper and the timeframes for responses, that the Reserve Bank is focused on concluding this process quickly. We question the Reserve Bank’s apparent urgency. New Zealand banks (in particular the four major banks) have consistently passed stress tests by the Reserve Bank (most recently in 20172), and no issues have been raised in relation to the ability of New Zealand’s banks to weather economic downturns by credit rating agencies or the IMF3. Costs/benefits need to be fully analysed We do not consider that it is sufficient for the Reserve Bank to conduct the cost/benefit analysis after it has received all submissions and immediately before any final decisions are made, as contemplated in the Reserve Bank’s 3 April 2019 background paper (the Background Paper)4. This will not give stakeholders any meaningful opportunity to comment on the factors the Reserve Bank has taken into account when making its assessment. The proposed changes to bank regulatory capital requirements are significant – by reference not only to the impact on New Zealand banks, their customers and the New Zealand economy but also when compared with the corresponding prudential requirements in other jurisdictions. For this reason we think it is necessary that a cost/benefit analysis is carried out as part of the consultation process. There have been historic concerns about how the Reserve Bank conducts and tests the cost/benefit analysis in respect of its preferred policy option and whether this is exposed to the same rigour as the cost/benefit analyses provided by stakeholders in support of alternative options. An example is the approach the Reserve Bank took to costing the outsourcing review5. In regards to the capital review, we note that market participants have commented that the Reserve Bank is overly optimistic about the benefits of its proposal and may have underestimated the potential costs. While there may be a self-serving element to some of these criticisms and there is rarely agreement among economists, we think that, for proposals of this significance, it is important that the Reserve Bank is transparent and has an open mind to feedback. Without this, we question whether stakeholders will have confidence in the Reserve Bank’s final decisions on the matter. 2 In its conclusions to the 2017 stress test of major banks the Reserve Bank stated “Outcomes from the test suggest that the four banks would be able to maintain capital levels above their minimum requirements during these scenarios”. 3 IMF described the major New Zealand banks as “resilient to a severe global economic downturn” – see “New Zealand: Financial System Stability Assessment” (IMF Country Report 17/110, May 2017), pgs 22 to 24. 4 Capital Review Background Paper: An outline of the analysis supporting the risk appetite framework – Page 1 Paragraph 5. 5 See, for example, the Cost of Proposals section in the Reserve Bank’s Regulatory Impact Statement – Revised Outsourcing Policy - February 2017.

RESERVE BANK: CAPITAL REVIEW PAPER 4: HOW MUCH CAPITAL IS ENOUGH? | 4 We note that a “success measure” for the Reserve Bank is to provide “....sound regulatory impact analyses of policies that the Reserve Bank intends to adopt…”.6 The level of capital should not be looked at in isolation The Reserve Bank appears to be looking at the level of capital banks need to hold in isolation rather than in conjunction with other relevant factors and levers. The financial crises which lead to bank failures often result from a range of underlying issues in financial markets. Resolving these, and limiting the potential impact of them, requires the use of a range of prudential tools. There is nothing in the information that the Reserve Bank has published to suggest that it has given any material consideration to the other tools in the prudential “toolkit” - such as open bank resolution (OBR), the core funding ratio and the liquidity coverage ratio. This is in contrast to other jurisdictions when central banks have considered adjustments to the capital adequacy framework, based on our review. In determining the capital levels under the European Capital Requirements Directive (CRD IV) and the Capital Requirements Regulations (CRR), for example, the European Commission also considered the management of liquidity risk, the definition of capital, counterparty credit risk and leverage ratios7. Likewise in respect of CRD V/CRR II, the proposed amendments provided a wide array of risk-reducing measures, including net stable funding requirements, total loss absorbing capacity (TLAC) requirements and a new market risk framework8. Separately, and notably, there is no indication in the Consultation Paper that the Reserve Bank has taken into account the impact that the outcome of the Reserve Bank Act Phase 2 review could have on the capital proposals. This seems strange to us, especially as the Consultation Paper was released within weeks of the Phase 2 discussion document. The Phase 2 review is considering a range of issues that will directly affect core elements of the Reserve Bank’s prudential “toolkit”, especially in areas where New Zealand has been out of step with international norms. These could have an 6 Reserve Bank of New Zealand Statement of Intent 2018 – 2021, pg 26, published June 2018. 7 See European Commission - Proposal for a Regulation of the European Parliament and of the Council on Prudential requirements for credit institutions and investment firms - 2011/0202 (COD) – see for example “2.2 Impact Assessment” page 5. 8 See European Commission - Adoption of the banking package: revised rules on capital requirements (CRR II/CRD V) and resolution (BRRD/SRM) – 16 April 2019. Note also that in relation to the Swiss “too big to fail” legislation, the proposal included strengthening the amount (and type) of capital (which was subsequently reviewed and amended, including by introducing a TLAC requirement), stricter liquidity requirements, better risk diversification and organisational measures to ensure the continuance of systemically important functions if a bank should fail.

RESERVE BANK: CAPITAL REVIEW PAPER 4: HOW MUCH CAPITAL IS ENOUGH? | 5 important bearing on considerations as to the optimal levels of capital for banks. The Phase 2 review will look at the effectiveness of the crisis management powers of the Reserve Bank. This may result in an overhaul of the OBR regime, and the introduction of alternative resolution regimes9. The outcomes of the Phase 2 review may also have a bearing on other prudential “toolkit” decisions that the Reserve Bank still needs to make (for instance, whether New Zealand introduces a leverage ratio, further policy developments in relation to the countercyclical buffer and the possible introduction of TLAC as an additional capital requirement). The Consultation Paper and the Background Paper considers in some detail the policy goals of the capital review, including weighing up the relative importance of the Reserve Bank’s current objectives of soundness and efficiency. However, the Phase 2 review may change the meanings assigned to, and the relative weightings of, the soundness and efficiency objectives, and may add new objectives, such as consumer protection or public confidence. Potentially even more significant are the questions in the Phase 2 review around depositor protection and the possible implementation of a depositor guarantee scheme. We consider that such a scheme would directly influence the level of risk depositors were prepared to be exposed to at a personal account level, and that it might lead the New Zealand public to consider that a one in 200 year risk profile for banks is too conservative. Further consideration of appropriate forms of capital instruments While we understand the Reserve Bank’s desire to ensure that any additional capital buffers should be of an appropriate quality, and that there are questions around the relative merits of going-concern and gone concern capital, we question the proposal to limit the form of the additional capital that is being proposed to Common Equity Tier 1 capital (CET1) only. In this context we consider that the decisions made by the Reserve Bank in earlier consultations on what types of financial instruments should qualify as bank capital should be reconsidered. CET1 capital may be simpler for the Reserve Bank to monitor and administer and, all other things being equal, we agree that simplicity should be preferred over complexity. But simplicity should not be the main determinant of what types of financial instruments qualify as bank capital. 9 See for example Bank of England (Martin Brooks et al) “Measuring the Macroeconomic Costs and Benefits of Higher UK Bank Capital Requirements” (Financial Stability Paper No. 35, December 2015). This noted that, inter alia, the beneficial effects of improvements in the UK’s resolution regime resulted in the authors arriving at a lower optimum capital ratio for UK banks than may otherwise be the case.

RESERVE BANK: CAPITAL REVIEW PAPER 4: HOW MUCH CAPITAL IS ENOUGH? | 6 In determining the regulatory requirements, the Reserve Bank should focus on what is the most optimal capital composition from the perspective of the efficiency of the financial system. Alternative capital instruments can also contribute to soundness by creating another group of stakeholders with a vested interest in a bank’s stability. Because they do not receive any upside return on their investment, they can be expected to be risk averse, with strong incentives to assert market discipline on the bank. We note that the Reserve Bank is out of step with the methodology being adopted in other major jurisdictions, including Australia, where considerable thought is being given to the optimal capital composition10. We are concerned that the Reserve Bank’s approach may be driven, at least in part, by constraints on its supervisory resources11. This seems a questionable justification, given that the cost of any increased resourcing by the Reserve Bank is likely to be significantly outweighed by the benefit of a more optimal capital position. Moreover, the resourcing and funding of the Reserve Bank is another of the key topics that the Phase 2 review is to consider – which, again, raises issues around the timing of the two reviews. 10 See APRA Discussion Paper - Increasing the loss-absorbing capacity of ADIs to support orderly resolution 8 November 2018 (especially Chapters 2 and 3) and the KangaNews article in March 2019 “APRA’s “door open” to ALAC developments”

• http://www.kanganews.com/news/9803-apra-s-door-open-to-alac-developments. 11 See for example the memorandum to the FSO from Financial Policy (Ian Woolford) dated 14 March 2016 entitled “A monkey on our back?” released by the Reserve Bank on 25 January 2019 as part of the background papers in relation to the review of capital.

Achieving monetary and financial stability in a growth economy requires dual control: SYSTEMIC LEVERAGE TARGETING (SLT) as well as INFLATION TARGETING Mechanism Monetary Policy Financial Policy Dimension of money addressed1 Price (with little impact on volume)2 Volume (with little impact on price)3 Target/benchmark Inflation rate Leverage of the financial system as a whole How defined? RPI/CPI Ratio of capital to assets across the system4 Government role Sets Target Sets Target Central Bank agencies responsible Monetary Policy Committee Financial Policy Committee Control lever Risk free rate of interest Capital ratio requirements for bank credit5 Agent of adjustment Open Markets Operations Department Prudential Regulation Authority Channel of adjustment OMO trades Basel II Pillar II instructions to institutions Determinant for adjustment Market Risk in specific asset classes What is adjusted? Asset prices Scaling factor/support factor for specific asset classes6 , or overall Capital Adequacy Ratio7 Outcome check Resulting inflation rate Capital in the system, as determined by stress testing/price of government debt8 Backup measurement Employment rate/economic growth rate Systemic risk Economic growth consists of the addition of new assets to the human economy. This is achieved by exploiting, or leveraging: raw materials; human labour; human ingenuity; or money – or usually a combination of these four things. Capitalism is the leverage of financial capital to create new assets. To the extent that this mechanism is undeveloped or fails, growth and increased prosperity will require greater leverage of labour or ingenuity. The optimal trade-off in human wellbeing is more difficult to achieve when there is a shortage of financial capital available for leverage, or when regulatory settings for leverage of capital are too restrictive. When regulatory settings for leverage of capital are insufficiently restrictive, the system can become unstable.

1 “The function of a Central Bank should be to regulate the volume and price of bank credit”, Committee on Finance and Industry, Report (Cmd 3897), para 303. 2 “The response of bank lending (to Bank Rate shocks) cannot be distinguished statistically from zero”, Bank of England Staff Working Paper 610, Aikman, Taylor and Bush 3 “tighter credit controls do not reduce prices”, ibid. 4 Precisely how this is calibrated should be an early focus for academic study, including the development of appropriate “weighting” for different equity holdings. 5 Does not preclude use of other macroprudential tools when appropriate. 6 The transition from Basel I to Basel II gave rise to a “scaling factor” of x 1.06; the EU’s CRR/CRD4 introduced a “support factor” or capital discount of x 0.7619 for SME lending. Changes to these factors should be set on a dynamic basis, according to asset class risk, which will translate through to the appropriate capital adjustment. 7 Leverage above target should trigger increases in required capital for asset classes carrying higher risk. Leverage below target could trigger a general reduction in required capital ratios, as implemented by the UK’s Financial Policy Committee in July 2016. 8 If a government chooses a high target, the economy will run “hotter”: so increased risk should increase the cost of government debt. A low target will mean a “cooler” economy, taking longer to pay back government debt – and so should also increase the cost. The optimal Systemic Leverage Target should be reflected in an optimal government debt price.

Chris Carswell I do believe that the capital of the banks needs to be increased to make the New Zealand banks safer. But I just wonder if giving the five years to do it is quick enough. We are reaching the end of one of the longest economic up cycles ever and it is difficult to see it going on for another five years without what could be a major recession. I also wonder that if there were problems with the Australian banks, would the contagion effect not kick in and the New Zealand banks would be in the same boat regardless of the level of capital. I have read the ‘bail in’ rules and while it states that the shareholders would take the first hit, I am unsure that this would happen in reality and that an argument would be made to keep them intact on the basis the shareholding makes up a major part of investment fund’s portfolios and that to wipe them would be catastrophic. Perhaps making it clear to the banks that there shareholding in the New Zealand banks would be wiped might make them act more responsibly. OIA s9(2)(a)

Clayton Dias Agree with the proposals to increase the amount of capital that banks need. OIA s9(2)(a)

From: Manji, Raf To: Capital Review Subject: Submission on bank capital Date: Thursday, 16 May 2019 5:13:56 PM Adrian Orr, Governor of the Reserve Bank of New Zealand, Wellington. Dear Adrian, In principle, I fully support the RBNZ’s proposal to increase bank capital requirements. There is no doubt that the deregulation of finance in the 1980s and beyond has boosted the global economy and brought greater prosperity through the world. However, the creation of a larger finance sector and an ability to leverage capital beyond responsible limits has led to repeated financial crises, unjustified house price increases and a risk taking mentality that has seen the risk/reward trade off heavily skewed towards taking risk, with the benefits flowing the private sector, and often, the costs being borne by the public sector. These costs have also been borne in our ecosystem, with those externalised costs now showing up in general environmental degradation and increasing climatic change. A clear example of the danger of loosening capital requirements, was the 2004 decision by the US Securities and Exchange Commission to allow investment banks to increase their leverage and self-regulate. https://www.nytimes.com/2008/10/03/business/03sec.html The results of this decision speak for themselves: a catastrophic collapse of the finance sector, leading to serious social impacts, which continue to this day in many countries. As a former investment banker, who has been involved in all aspects of risk and capital, from analysis to trading, I can say without any doubt that the finance sector has enjoyed a free ride at the expense of society in general, particularly to those who fund capital hard to access, and rely purely on wages. This proposed shift in bank capital expectations sends the message that banks operate under a social license. The profits generated over the last 20 years speak volumes. The risk that now sits in the system in the form of household debt, the disconnection of incomes and house prices (supported by lower interest rates) and the focus on property as collateral for business investment is of serious concern. Whilst this is not an easy conversation to have, I applaud the Reserve Bank for having the courage to address this issue, and realign the banking sector onto a more sustainable path. I would have liked to present a fuller submission but due to work and time constraints, I have not been able to do so. That being the case, I would like to speak to my submission should the opportunity be available. Kind regards,

Raf Raf Manji, Chair of the Finance and Performance Committee, Councillor for Waimairi, Christchurch City Council.

---

This electronic email and any files transmitted with it are intended solely for the use of the individual or entity to whom they are addressed. The views expressed in this message are those of the individual sender and may not necessarily reflect the views of the Christchurch City Council. If you are not the correct recipient of this email please advise the sender and delete. Christchurch City Council http://www.ccc.govt.nz

---

OIA s9(2)(a)