towards a separation of deposit and investment banking activities

Towards a separa+on of deposit and investment banking ac+vi+es

Jan Pieter Krahnen

Center for Financial Studies at Goethe University SAFE, CEPR

18 April 2013 Towards the Banking Union: Open Issues Roundtable at College of Europe, Bruges

Agenda

I.  LIIKANEN PROPOSAL

II.  BACKGROUND ON SEPARATION -  Trading, banking, and risk taking. -  Market making and banking.

III.  ALTERNATIVE MODELS OF SEPARATION -  Comparing Volcker, Vickers, Liikanen.

IV.  ISSUES FOR DISCUSSION AND RESEARCH

2

Proposals need a narra<ve

Crisis narra+ve §  Consistent economic explana+on of why the crisis happened, and how it

progressed. §  Based on available academic and policy research, §  As well as expert interviews on European bank business models.

3

Proposals and their narra<ves US (Volcker): Excessive risk taking (through propietary trading) à Spin off prop trading from deposit taking UK (Vickers): Excessive risk taking (through

investment or commercial banking) affects economy‘s core financial services. à Ring-‐fence core services & allow other banking ac+vi+es to default (eliminate TBTF for ring-‐fenced ins+tu+ons).

EU (Liikanen): •  Systemic risk as the key challenge for preven+on and

interven+on. •  Overcoming too-‐big-‐to-‐fail (TBTF) à  Two main sugges+ons:

-  Enabling Resolvability (despite complexity) -  Revitalizing Market Discipline via true Private Liability

(i.e. equity and debt) 4

Liikanen Proposal: basic approach

Facilita+ng resolvability of banking ins+tu+ons

No recourse to taxpayers’ money (pushing systemic risk back to tail of

distribu+on)

Mandatory issuance of junior bank debt, held outside banking

system (First-‐in-‐line debt) Separa+on of trading ac+vi+es

from universal banking

Re-‐introducing market discipline into banking system

Objec+ve

Strategy

Instruments

Result

5

A common misunderstanding

§  Instruments to be bailed-‐in •  Cocos, Debt write-‐down, With pre-‐determined trigger, With pre-‐

determined conversion formula, respec+ng seniority

§  By some commentators, “bail-‐in bonds” were misread: •  It has been insinuated that newly created „bail-‐in bonds“ is the one

and only instrument to be bailed-‐in by the supervisor. •  All other debt instruments, in this interpreta+on, would con+nue to

benefit from a governmenbt guarantee. •  Needless to say, this puts our proposal upside down.

§  ALL DEBT IS BAIL-‐IN-‐ABLE, subject to seniority, and regardless of its label.

§  Change label now: à First-‐in-‐line bonds à Anchor bonds à No-‐mercy bonds

6

Liikanen Report: proposals

1)  Mandatory separa,on of proprietary and significant other trading ac+vi+es §  Addi+onal separa+on of other ac+vi+es condi+onal on recovery and

resolu+on plan.

2)  Amendments to the use of bail-‐in instruments as a resolu+on tool.

3)  Review of capital requirements on trading assets and real estate finance (increasing risk weights).

4)  Strengthen banks’ governance and control (aligning incen,ves).

7

First-‐in-‐line/Bail-‐in-‐ability

Ensuring bank resolvability without

recourse to TPM

All banks: Mandatory issuance of designated first-‐in-‐line debt.

May be subs+tuted by all-‐equity. Does not imply other debt to be bail-‐out debt.

Loss absorp<on First-‐in-‐line bonds are available for write-‐down (long-‐term investors can distribute bail-‐in losses over many periods).

Or conversion into equity (coco bonds), the Swiss example.

Incidence: First-‐in-‐line bonds are priced accordingly (high coupon compensates for high expected loss).

First-‐in-‐line layer may have tranches of different seniority.

Credibility: First-‐in-‐line bond investors must be non-‐banks, limi+ng contagion risk (eliminates the rescue impera+ve for the state).

May be subs+tuted by a 1250% risk weight.

8

Agenda


II.   BACKGROUND ON SEPARATION -  Trading, banking, and risk taking. -  Market making and banking.



9

Theore<cal perspec<ve on banking and trading

Boot/Ratnovsky 2012

Banking §  Rela+onship business §  Not scalable §  Safe, if lending book is

large and diversified. §  Franchise value via long

term pricing.

Trading §  Transac+onal business §  Scalable, with decreasing

returns to scale. §  High risk-‐return strategy

possible. §  No franchise value.

Possible problems of banking and trading interac<on §  Time inconsistency Banks may allocate too many resources to trading, undermining rela+onship business.

§  Risk shiling Trading may be used to increase risk, to the benefit of shareholders.

10

Policy implica<ons

§  Rela+onship banks are tempted to “use their balance sheet” for trading ac+vi+es.

§  Segrega+ng resources within the universal bank (firewalled subsidiaries)...

•  …solves +me inconsistency problem: because capital remains in rela+onship business.

•  ...solves risk shiling problem: because trading funding is fully risk-‐sensi+ve. Not subsidized.

§  Volcker, Vickers, Liikanen. •  All agree on a de-‐minimis rule, •  Because banking-‐trading interac+on is beneficial if carried out

at moderate level.

11

Interview evidence on banking and trading

Tradi<onal investment banking §  Broker-‐dealer, ac+ve in a defined set of markets. E.g., M&A,

corporate loans, foreign exchange products, bond issues, IPOs. §  Personal rela+onships and repeat interac+on maner. §  Bank matches demand and supply. §  Moderately-‐sized balance sheet – leverage is a by-‐product of

the brokerage role. §  Income source mostly fees and commissions. §  If spread income, then from prop trading desk.

Modern investment banking (MIB) §  Upgrading into risk-‐engineered, high value-‐added services of a

central counterparty. §  Real synergies bw. investment banking and commercial/retail bkg. §  “Plaporm model”

12

Financial produc<on plaSorm

PlaSorm explained §  Plaporm consits of a wide array of basic products (e.g. op+ons, swaps,

CDS, bonds) and processes (structuring, syndica+on, internaliza+on, neqng, electronic crossing networks, dynamic hedging using factor models)

§  MIB: from match-‐finder to provider of counter-‐party services. §  Consider bespoke financial product, e.g. a corporate credit risk of a

par+cular quality.

PlaSorm business and balance sheet growth §  With capacity in place to deliver func+onal (fin. eng.) services, the bank

can start to leverage its generic customer business. §  Building on order flow of its customers, the bank can subs+tute for the

other side of many transac+ons. §  It makes the market, entering exposures into its trading book. §  Balance sheet reflects accumula+on and holding of inventories

13

PlaSorm strategy and risk

Professional literature §  :“flow monsters”, banks with large flows of customer business that

facilitate market making. §  Plaporm costs are largely fixed, sugges+ng increasing returns to scale. §  Flow monsters turn into stock monsters, banks with large balance sheets. §  High leverage is by-‐product of market making, not intended risk strategy. §  So is risk taking.

Implica<ons §  Not always easy to prove the case for market making (e.g. factor hedging)

-‐-‐> a sequence of prop trades. §  Own capital is at risk, although no prop trading. §  By connec+ng financial ins+tu+ons to a wider financial network, hybrids

may contribute to systemic risk (Chow/Sur+ 2011)

14

Cau<ous policy implica<ons

On bans and ring-‐fences

§  Banning generic prop trading (Volcker) •  Does not erase flow prop trading, unless market making is defined

very restric+vely (i.e. high-‐liquidity markets)

§  Ring-‐fencing commercial and retail banking (Vickers) •  Investment and commercial banking are intertwined, even if there

is no specula+on. •  May affect MIB severely, because of exposure limits

§  Banning prop trading and market making (Liikanen) •  May affect MIB severely, because of exposure limits.

15



III.   ALTERNATIVE MODELS OF SEPARATION -  Comparing Volcker, Vickers, Liikanen.


Agenda

16

Overview

Market Making

Underwri+ng

Selected banking ac<vi<es

Commercial banking Investment banking

Retail banking (services to individuals

and SMEs)

Wholesale banking (services to financial

ins<tu<ons and corporates)

Overdrals & Lending

Deposit taking

Asset & Wealth Management

„Narrow Investment Banking“

(services to financial ins<tu<ons and corporates)

Ancillary and proprietary services

(services to the bank)

Overdrals & Lending

Deposit taking

Payment func+ons

Investment services

Payment func+ons Repos

Hedging

Prop. Trading

Advisory

Sales & Trading

Trade finance

Altern. investments (hedge funds)

Investments in gov. securi+es

Prime brokerage

17

Glass Steagall

Market Making

Underwri+ng





Overdrals & Lending

Deposit taking






Overdrals & Lending

Deposit taking

Payment func+ons

Investment services


Hedging

Prop. Trading

Advisory

Sales & Trading

Trade finance



Prime brokerage

Permined

Prohibited

18

Volcker

Market Making

Underwri+ng




and SMEs)



Overdrals & Lending

Deposit taking






Overdrals & Lending

Deposit taking

Payment func+ons

Investment services


Hedging

Prop. Trading

Advisory

Sales & Trading

Trade finance



Prime brokerage

Permined

Prohibited

Permined under condi+ons 19

Volcker Rule

§  The Volcker Rule prohibits any banking en,ty and its affiliates (any type of credit union or insured depository ins+tu+on) from sponsoring or inves,ng in any hedge fund or private equity fund, as well as any other type of private investment fund (structured, VC, real estate etc.)

§  The Volcker Rule also prohibits engaging in proprietary trading, which is defined as short-‐term trading (the purchase and sale of financial securi+es, encompassing all stocks, bonds and deriva+ves) with the intent to profit from the difference between the purchase and sales price

20

Volcker Rule -‐ Exemp<ons

§  De minimis „3 Percent“ rule: •  Bank are allowed to invest up to 3 percent of their Tier-‐1 Capital in

funds as long as they do not own more than 3 percent of each invested fund

§  Exempted from proprietary trading prohibi+ons are: •  Municipal Bonds •  Market making, clearing and underwri+ng ac+vi+es •  Hedging and liquidity management ac+vi+es •  Repos •  Any trades conducted on behalf of customers

21

Volcker Rule – Default Rule and Compliance

§  All affected banks must file reports with their regulatory authori+es to show compliance with the rules

§  All permined ac+vi+es under Volcker are subject to a default prohibi+on rule: none of the permined ac+vi+es are allowed if •  They are subject to excessive risk-‐taking (more than would be

appropriate to provide financial intermedia+on services) •  Include compensa+on systems which primarily reward proprietary

risk-‐taking •  The trading ac+vi+es have the bank pay more fees than they obtain

through the ac+vi+es •  The trading ac+vi+es show high earnings vola+lity and earn either

very small or very large revenues per unit of risk taken

22

Volcker Rule -‐ Challenges

§  Permined versus prohibited ac+vi+es? •  What is „hedging“? •  When is prop trade a leg in a market making transac+on, when is it not?

§  What are effects on market liquidity, risk premia? •  Thakor (2011) predicts drop in liquidity for smaller markets, increase in

cost of capital, lower real investment, less efficient risk management.

23

Vickers

Market Making

Underwri+ng




and SMEs)



Overdrals & Lending

Deposit taking






Overdrals & Lending

Deposit taking

Payment func+ons

Investment services


Hedging

Prop. Trading

Advisory

Sales & Trading

Trade finance



Prime brokerage

Permined

Prohibited for the ring-‐fenced en+ty, but allowed for the non-‐ring-‐fenced en+ty.

Permined for the ring-‐fenced en+ty as service to corporates, but prohibited as service to financial ins+tu+ons.

Services to non-‐European customers

24

Vickers – General idea

§  The commission’s recommenda+ons on structural reform are to partly separate UK retail banking services from global wholesale and investment banking services, the so-‐called “retail ring-‐fencing”.

§  The idea behind this separa+on is to credibly restrict public guarantees to ring-‐fenced banks that perform vital banking services, and thereby to take away incen+ves for excessive risk-‐taking of non-‐ring-‐fenced banks. •  Ring-‐fencing would help “insulate UK retail banking from global shocks”

and ensure the supply of credit in the economy. •  “Retail deposits – now around £ 1 trillion – would fund loans to

households and businesses in the domes+c economy, not investment banking.” (Vickers, 2010).

25

Vickers – Details of the ring-‐fence

§  Mandated services for the ring-‐fenced en+ty: Taking of deposits from and providing overdrals to individuals and SMEs

§  Prohibited services for the ring-‐fenced en+ty: •  Investment banking ac+vi+es such as deriva+ves, debt and equity

underwri+ng and inves+ng and trading in securi+es. •  Commercial banking services resul+ng in exposures to financial companies. •  Banking services to non-‐European customers

§  The ring-‐fence is flexible in allowing banks to place other ac+vi+es such as lending to large domes+c corporate and trade finance inside or outside the ring-‐fence.

§  Permined ancillary services for the ring-‐fenced en+ty: •  Ac+vi+es necessary for the efficient provision of mandated services, i.e.,

typical treasury func+ons such as risk management (e.g., interest rate hedging through deriva+ves) and liquidity management.

§  Specific pruden+al safeguards of the ring-‐fenced bank •  More equity, stricter leverage limits, bail-‐in, depositor preference, primary

loss absorbing capacity …

26

Vickers – Links between fenced and non-‐fenced en<ty

§  Ring-‐fenced banks should be self-‐standing or subsidiary companies in wider banking groups.

§  Legal and opera+onal links •  The ring-‐fenced en+ty should ensure that it can be isolated from

the group in a few days and can con+nue to provide services.

§  Economic links •  The ring-‐fenced bank's rela+ons with other parts of the group

should take place on a third party basis. •  The ring-‐fenced bank should not be dependent on the group's

con+nued financial health for its solvency or liquidity.

27

Vickers – no “de minimis” rule

§  In the Vickers Report, no “de minimis” limit was recommended under which the ring-‐fencing requirement would not apply. •  “Any fixed costs associated with ring-‐fencing would be propor+onately

greater for smaller banks. However, complex small banks could s+ll pose significant resolu+on challenges, an exemp+on could confuse consumers, and the risk of contagion from financial markets to the retail banking system would remain if there were a large number of small banks opera+ng below some de minimis limit.“ (Vickers Final Report, p. 39)

§  However, the UK Government Response states that the case for “de minimis” exemp+ons from ring-‐fencing, in par+cular for very small firms, should be reviewed.

28

Vickers -‐ Challenges

§  Implementa+on challenges similar to those from the Volcker rule may emerge

§  Clarify which risk-‐management ac+vi+es are allowed within the ring-‐fence.

§  To the degree that risk-‐management of the ring-‐fenced en+ty is done via the non-‐ring-‐fenced en+ty of the bank, this creates contagion risk.

§  A bank holding structure with a ring-‐fenced and a non-‐ring-‐fenced en+ty may create incen+ves to operate the ring-‐fenced en+ty at minimum capital ra+os.

29

Liikanen

Underwri+ng




and SMEs)



Overdrals & Lending

Deposit taking






Overdrals & Lending

Deposit taking

Payment func+ons

Investment services


Advisory

Sales & Trading

Trade finance


permined

prohibited for concerned party (i.e. beyond de-‐minimis) 30

Market Making

Prime brokerage

Hedging

Prop. Trading

Altern. Investments (hedge funds)

Summary of Recommenda<ons

Separa+on of trading (prop trading and

market making)

Separate legal en+ty (broker-‐dealer), holding structure, restricted joint liability, exposure limits apply.

Stand-‐alone funding of trading ac+vity avoids implicit subsidiza+on.

Resolu+on is facilitated.

Generous de minimis rule applies

Universal banking model (commercial + investment) remains largely untouched.

31

De-‐minimis rule

Mandatory Separa+on only if the concerned ac+vi+es amount to a significant share of a bank business.

Assessment to be completed in two stages:

1)  Iden+fy banks whose assets “held for trading “ and “available for sale” exceed: a. a rela+ve threshold of 15-‐25% of total assets; or b. an absolute threshold of EUR 100bn

3)  Supervisors decide case-‐by-‐case on the basis of the assets to which the separa+on requirement applies. The threshold is to be calibrated by the Commission.

32

Other features

§  Higher capital charges were discussed as an alterna+ve to separa+on. §  Broker-‐dealer and universal bank under one holding. §  Broker-‐dealer has own funding, equity and debt §  Cross guarantees uni-‐direc+onal (from TB to UB, but not reverse) §  Prop trading not prohibited for holding company (at TB, not at UB) §  Two roads to separa+on

•  Avenue 1: first, higher capital buffer (not risk-‐weighted) for banks with large trading book. Then RRP. Second, if RRP not accepted by supervisor, separa+on is mandatory.

•  Avenue 2: Immediate separa+on, upon crossing a defined threshold (e.g., trading assets)

33

Proposal by HLEG Banking Group in comparison

pro

Prop trading+ Market making

Investment and commercial banking (prop trading, mkt making, large corporate lending)

”Volcker Banking Group”

”Vickers Banking Group”

”HLEG Banking Group”

Investment and commercial banking

Prop trading+ Mkt making

Investment and commercial banking

Swaps push-‐out

Retail banking with higher capital requirements

34





Agenda

35

What will be the market response?

§  Market response is a general equilibrium issue.

§  Emergence of broker-‐dealers – as independent arms of Sifis. •  Need to fulfill capital requirements as well, including bail-‐in (first-‐in-‐

line) debt.

§  Will smaller ins+tu+ons (below-‐de-‐minimis) grow their trading business, taking over from the broker-‐dealers? •  Probably not, assuming fixed costs in trading to be significant. •  Rather, and for the same reason, broker-‐dealers may take over

business from smaller ins+tu+ons.

36

Will there be less liquidity?

§  What happens to plaporm strategy of “flow monsters”? •  Possibly, financial engineering advisory will con+nue as before, now

provided by the broker-‐dealer, only execu+on of resul+ng net posi+ons is outsourced to one/more broker-‐dealers.

•  Thus, conceivably, plaporm strategy can be largely maintained. •  However, broker-‐dealer has to pay funding costs commensurate with

its stand-‐alone risk. §  Profitability of market making-‐cum-‐prop trading likely to shrink

•  (General equilibrium adjustment is complicated to forecast, however)

§  Effect on trading volume and liquidity •  Thakor (US Chamber of Commerce, 2011): Volcker rule will reduce

liquidity, increase spreads, lower credit volume, lower economic growth.

•  Does not consider general equilibrium repercussions, e.g. new entrants (broker-‐dealers), new price level, new market models.

37

Will there be less risk transfer?

§  Securi+za+on of the loan book was the early success story of a coopera+on between commercial and investment banking. Will this be made impossible?

§  Probably no material change, if one allows structuring exper+se to be outsourced (to the ring-‐fenced trading house, for instance).

§  Consider the joint effect of separa+on and bail-‐in debt issues. •  Bail-‐in debt transfers the most junior debt layer in the bank balance

sheet to private investors outside the banking system, making it more resilient against shocks.

•  More risk transfer than before the crisis likely.

38

What about 2-‐<er banking model? (1)

§  Money center banking •  Within-‐group borrowing and lending not affected by separa+on. •  Single counterparty exposure limits apply.

-  Pushes Groups to consolidate (as within-‐Group exposures do no longer maner).

§  Central deriva+ves and trading facility of the Group •  Within exposure limits: unaffected by separa+on •  Beyond exposure limits: exact same transac+ons executed through

other market par+cipants -  ‘Other market par+cipant’ could be a white-‐labeling, compe+ng bank. -  ‘Other market par+cipant’ may be another daughter, in or out of mul+ bank holding

company (no longer in US: Intermediate Holding Company under DFA for Foreign Banks).

39

What about 2-‐<er banking model? (2)

§  Effect of separa+on regula+on on business model •  Level 1 bank (retail): no material effect, just different provider in

market. •  Level 2 bank (wholesale): reduced transac+on volume with Group

members, but poten+al for increased transac+on volume with members of other Groups (gen. eq. effect)

§  Effect of bail-‐in regula+on on business model •  Depends on Group status (consolidated or not)

-  If fully consolidated: bail-‐in debt requirement applies at Group level only. -  If not fully consolidated: bail-‐in debt requirement applies at the individual

bank level -‐-‐> absolute volume increases (double-‐coun+ng). -  If not fully consolidated: Group-‐internal deposit insurance schemes are

affected as well, if the scheme covers all liabili+es, as in the case of a ‘Ins+tutssicherung’ (savings banks, Germany).

40

Trading and banking -‐ a research request

§  Not much empirical work exists on the role of securi+es and deriva+ves trading in today’s banking markets.

•  Interviews are not enough. •  Clinical studies needed: understanding the business model of large,

interna+onal banks, and the role played by trading ac+vi+es (plaporm strategy)

•  More generally, more should be known about the welfare effect of increased trading in world capital markets. The risk we have seen – where is the benefit?

•  What are the expected costs of curtailing trading for society?

§  What are investor incen+ves when banks rely on mul+-‐+er capital structure (as opposed to simple debt-‐equity schemes)?

§  Similarly, understand role of interbank secured (repo) and unsecured market.

41





V.  APPENDIX

Agenda

42

Acharya, V., Pedersen, L.H., Philippon, T. and Richardson, M. (2010), “Measuring Systemic Risk”, unpublished working paper

Acharya, V., Schnabl, P. and Suarez, G. (2011), "Securi+za+on Without Risk Transfer", forthcoming, J. Financial Econ.

Acharya, V., Shin, H.S. and Yorulmazer, T. (2011), "Fire-‐sale FDI", forthcoming, Korean Economic Review

Adma<, A., DeMarzo, P., Hellwig, M., Pfleiderer, P. (2011), “Fallacies, irrelevant facts, and myths in the discussion of capital regula+on: why bank equity is not expensive”, Stanford GSB working paper.

Adrian, T. and Shin, H.S. (2010), “Liquidity and Leverage”, Federal Reserve Bank of New York Staff Reports

Boot, A. and L. Ratnovsky (2012), “Banking and trading”, Working paper, University of Amsterdam.

Borio, C. (2006), “Monetary and pruden+al policies at a crossroads? New challenges in the new century”, BIS Working Papers, No 216

Bruche, M. and Suarez, J. (2011), “Deposit insurance and money market freezes”, J. of Monetary Econ. 57, 45–61

Brunnermeier, M.K., (2009), “Deciphering the Liquidity and Credit Crunch 2007–2008”, Journal of Economic Perspec+ves, Volume 23, Number 1, p 77–100

Brunnermeier, M.K., Dong, G. and Palia, D. (2012), “Banks’ Non-‐Interest Income and Systemic Risk”, unpublished working paper

Demirgüç-‐Kunt, A. and Huizinga, H. (2010), “Bank ac+vity and funding strategies: The impact on risk and returns”, Journal of Financial Economics 98, p. 626–650

Duffie, D. (2011), “Systemic Risk Exposures: A 10 by 10 by 10 Approach”, NBER WORKING PAPER SERIES, Working Paper 17281

Fahlenbrach, R. and Stulz, R.M. (2011), “Bank CEO incen+ves and the credit crisis”, Journal of Financial Economics 99, p. 11–26

References , part 1

43

French, K.R., Baily, M.N., Campbell, J.Y., Cochrane, J.H., Diamond, D.W., Duffie, D., Kashyap, A.K., Mishkin, F.S., Rajan, R.G., Scharfstein, D.S., Shiller, R.J., Shin, H.S., Slaughter, M.J., Stein, J.C. and Stulz, R.M. (2010), “The Squam Lake Report -‐ Fixing the Financial System”, Princeton University Press: Princeton and Oxfor

Gorton, G. (2012), “Misunderstanding Financial Crises: Why We Didn’t See One Coming”, Oxford University Press, forthcoming

Gorton, G. and Metrick, A. (2011), “Securi+zed banking and the run on repo”, forthcoming J. Financial Economics

Haldane, A.G. (2009), “Rethinking the financial network”, Speech by Mr Andrew G Haldane, Execu+ve Director, Financial Stability, Bank of England, at the Financial Student Associa+on, Amsterdam, 28 April 2009.

Hanson, S.G., Kashyap, A.K. and Stein, J.C. (2011), „A Macropruden+al Approach to Financial Regula+on”, Journal of Economic Perspec+ves, Volume 25, Number 1, p. 3–28

Hoshi, T. (2011), “Financial Regula+on: Lessons from the Recent Financial Crises”, J. Econ. Lit. 49, 120–128.

Issing, O. (2009), “Why a Common Eurozone Bond Isn’t Such a Good Idea”, Center for Financial Studies White Paper No. III

Mishkin, F.S. (2011), „Over the Cliff: From the Subprime to the Global Financial Crisis“, Journal of Economic Perspec+ves, Volume 25, Number 1, p. 49–70

Pozsar, Z., Adrian, T., Ashcrak, A. and Boesky H. (2012), “Shadow Banking”, FRB New York Staff Reports

Rajan, R.G. (2006), “Has Finance Made the World Riskier?”, European Financial Management, 12, p. 499–533

Reinhart, C.M. and Rogoff, K.S. (2011), “A Decade of Debt”, NBER Working Paper 16827

Tirole, J. (2011), “Illiquidity and All Its Friends”, Journal of Economic Literature 2011, 49:2, p. 287–325

References , part 2

44

towards a separation of deposit and investment banking activities

Documents