Whats New in Basel III

BASEL III: WHAT’S NEW?BUSINESS AND TECHNOLOGICAL CHALLENGES

SEPTEMBER 17, 2010

By Rustom Barua, Fabio Battaglia, Ravindran Jagannathan,Jivantha Mendis and Mario Onorato

2

Basel III: What’s New?Business and Technological ChallengesSeptember 17, 2010

Table of Contents1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2. Liquidity Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2.1. The Regulatory Effort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

2.2. The New Liquidity Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

2.3. The New Liquidity Ratios:Tasks and Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2.3.1. Insufficiency of Standardized Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2.3.2. Building Differentiated Incentives to Traditional Banking vs. Speculative Trading . . . . . . . . . . . . . .10

2.3.3. Accounting for Bank’s Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11

2.3.4. Need to Raise New Capital . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11

2.3.5. Securitization Disruption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12

2.3.6. Raising New Medium/Long-Term Finance (NSFR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13

2.3.7. Distorting Bond Markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14

2.3.8. Reshaping Interbank Deposit Markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14

2.4. Survival Horizon Models: Optimizing the Liquidity Buffer in a Comprehensive Risk Management Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

2.4.1. The Relationship between the Basel Ratios and Bank-Specific Survival Horizon Models . . . . . . . .15

2.4.2. Calculating the Amount of the Liquidity Buffer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16

2.4.3. Optimising the Liquid Asset Buffer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17

3. Proposals Regarding Capital . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18

3.1. Capital Base . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18

3.2. Risk Coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19

3.2.1. Addressing General Wrong-way Risk – Stressed EEPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19

3.2.2. Capturing CVA Losses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20

3.2.3. Specific Wrong-way Risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21

3.2.4. Higher Risk Weights for Financial Institutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21

3.2.5. Increase Margin Period of Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22

3.2.6. Preclude Downgrade Triggers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22

3.2.7. Collateral Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23

3.2.8. Central Counterparties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23

3.2.9. Stressed PDs for Highly Leveraged Counterparties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23

3.2.10. Stress Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24

3.2.11. Back Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24

3.2.12. Reduce Reliance on External Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24

3.3. Leverage Ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25

3.4. Counter-Cyclical Capital Buffers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27

3.5. Systemic Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31

4. Implementation Timelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32

5. Business impact and challenges: exploring the interplay between Liquidity and Capital . . . . . . .32

5.1. Exploring Interconnections and Trade-Offs between Capital and Liquidity . . . . . . . . . . . . . . . . . . . .34

5.2. Misunderstanding How Liquidity Risk and Capital are Connected . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35

6. Technology Direction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36

7. Conclusion: Moving Towards a Holistic System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39

REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40

3


1. IntroductionAn extensive effort is underway to strengthen the financial sector and make banks and other institutionsmore resilient in the face of unexpected stress.The hope is that any future crisis will not lead to governmentsagain being forced to spend billions of dollars of taxpayer’s money saving the banking system.

In terms of regulatory requirements, this effort has been concentrated in proposals envisaging threeareas where constraints are being substantially overhauled: regulatory capital, liquidity and leverage.These proposals were summarized in two Consultation Papers issued by the Basel Committee on BankingSupervision (BCBS) In December 2009:

• Strengthening the Resilience of the Banking Sector, dealing with regulatory capital and leverage.

• International Framework for Liquidity Risk Measurement, Standards and Monitoring, addressingliquidity requirements.

The urgency of the tasks was expressed by Mario Draghi, the Chairman of the Financial Stability Board, inhis Letter to the gathering of G20 world leaders in Toronto. Published on June 27th, 2010, it read:“It willbe important that Leaders support calibration of the new capital, liquidity and leverage standards toa level and quality that enable banks to withstand stresses of the magnitude experienced in this crisis,without public support. The quality and amount of capital in the banking system must be significantlyhigher to improve loss absorbency and resiliency.We should provide transition arrangements that enablemovement to robust new standards without putting the recovery at risk, rather than allow concerns overthe transition to weaken the standards”.

On July 26, 2010 the Bank for International Settlements (BIS) announced that the Group of Governorsand Heads of Supervision, the oversight body of the Basel Committee, had reached a broad agreementon a capital and liquidity reform package. This resulted not only in a significant easing of the rulescompared to the first draft, but more importantly in a substantial delay in their effective implementation.For the leverage ratio and the net stable funding ratio, which had been concerning banks most, transitionperiods were established such that the new rules will not come into force until 2018. Jean-Claude Trichet,President of the European Central Bank and Chairman of the of Governors and Heads of Supervision, madeclear that this delay is aimed at avoiding the possibility of the new constraints hitting the global economywhile a difficult recovery is in course:“ We will put in place transition arrangements that ensure the bankingsector is able to support the economic recovery”1.

The BIS communiqué stated that the Governors and Heads of Supervision had taken account of the resultsof the quantitative impact study undertaken by the Basel Committee to assess the potential impact onbank profitability and the broader economy of the new rules.The results of this study will be published bythe Committee later this year.

At its 12 September 2010 meeting, the Group of Governors and Heads of Supervision fully endorsed theagreements reached on 26 July 2010. These capital reforms, together with the introduction of a globalliquidity standard, deliver on the core of the global financial reform agenda and will be presented to theG20 Leaders summit taking place in Seoul in November.

1 Bank for International Settlements,The Group of Governors and Heads of Supervision reach broad agreement on Basel Committee capital and liquidity reformpackage, Press Release, 26 July 2010.

4


The agreement was widely seen by the market as good news for banks, and bank shares worldwide spiked.Banks will still be allowed to recapitalize through retained earnings rather than fresh capital and will pushback into the future the downwards pressure on profits that would result from the obligation to holdgreater amounts of capital, liquid assets and medium/long-term debt. However, concern was expressed bycommentators as to the effectiveness of such a long transition in view of system protection.We could alsoquestion whether European supervisors have fully taken advantage of lessons from the crisis.The above“win”on July 26 for banks came just days after the European Central Bank released the results of the stresstests conducted on a significant sample of European banks.The tests showed that most of these would becapable of withstanding a significant and protracted stress. However, the tests focused on capital levelsonly and did not test banks for liquidity risk.This persistence of a “silo”approach to risk management is, inour view, to be seen as a weakness. There is significant evidence from the crisis that interdependenceamong risks cannot be dismissed and that unexpected fallouts in terms of liquidity can put financialinstitutions at extreme risk.We will outline in the last part of this document that the “silo”approach shouldbe completely overcome in favour of an integrated view of different risk types that duly considersinterdependencies among risks.

This paper will focus on the current version of the new Basel requirements on capital, liquidity and leverageas amended after the 26 July communiqué and ratified on September 12, 2010. It will analyse implications,issues and interconnections between them and discuss some of new trends in best practice of banks’ riskmanagement and capital optimization that are likely to emerge as a result.

The enhanced set of rules has been widely referenced in the industry as “Basel III”. When the Basel II Accordwas finalized in 2004 the assumption was that the overall quality and quantity of capital was sufficient,but the BCBS wanted to make the regulatory capital measure more risk sensitive. Current discussions onBasel III intend to achieve better quality capital as well as increasing the amount of capital.

In this document we will first describe the liquidity risk rules and potential shortcomings in Basel III.In section 3 we describe changes to capital requirements, with special emphasis on risk coverage, leverageratio and countercyclical capital buffers. In section 4 we summarize the implementation timelines asspecified in the September 12,2010 press release.Section 5 describes the potential business impact for globalbanks and explores the interplay between capital and liquidity. Section 6 is our view of how enterpriserisk technology will evolve over the next few years.We conclude with section 7 on how business processesand systems are moving towards a holistic, integrated risk management framework.

5


2. Liquidity Risk After being neglected for decades, liquidity risk has suddenly taken centre stage thanks to the financial crisis.As a result a continuous flow of new best practice guidance and supervisory requirement documents haveemerged over the last couple of years.

Such a stringent approach to liquidity risk supervision is indeed rather new in the regulatory framework.In both Basel I and Basel II, liquidity risk received only limited attention. The entire Basel framework onlylooked at the asset side of the balance sheet. Risks arising from the liability side (including liquidity riskalongside other risks, such as interest rate risk of the banking book), for instance, are not subject to anyregulatory capital requirement.They are instead disciplined under Pillar 2, whereby banks are required toundertake the ICAAP (Internal Capital Adequacy Assessment), i.e. a calculation of the amount of capital(called internal capital) they deem sufficient to support all their risks. Pillar 2 requires that the ICAAPinclude liquidity risk. However, this provision has resulted in an inconsistency: after years of sterile debateon the possible methodologies for calculating internal capital for liquidity risk, it has been generallyaccepted that capital is not a suitable mitigant for liquidity risk. As a result, the current Basel II frameworkdoes not in effect address liquidity risk.

At the root of this construction stood a very fundamental assumption, that a bank would always becreditworthy as long as asset quality was preserved. In other words, provided the quality of assets wasgood enough then a bank would always find finance at fair prices, for virtually any amounts.

This assumption proved completely wrong when the crisis erupted and entire liquidity channels suddenlydried up, such that even institutions with high ratings and excellent asset quality found themselves troubleas a result of liquidity mismatches. This phenomenon grew to systemic proportions since many in theindustry had been massively leveraging maturity mismatches between assets and liabilities as a keycomponent of an extremely profitable business model.

Liquidity risk originates from the mismatch between the timings of cash inflows and outflows. As such, itis fundamentally inherent to the banking business. In fact, one of the key functions of the banking industryin a modern economic system is to allow the reallocation of financial resources from the liquid sectors(those which have excess financial resources to invest) to the illiquid ones.This entails two consequences:

1. The banking industry is necessarily exposed to a maturity mismatch. Typically, the term onwhich liquid operators are ready to invest their liquidity is shorter than that on which illiquidoperators are willing to borrow. While reallocating financial resources from one sector to theother, the banking system bears such mismatch of maturities in the form of liquidity risk.

2. The banking industry is a leveraged one. Its business is borrowing money from excess sectorsand lending it to sectors in need. Banks inherently work on others’ money. Obviously, a highleverage boosts the impact of any liquidity problem, both on an individual and a system basis.

As a result, regulators cannot aim to remove mismatch liquidity risk from the system. One individual bankcould theoretically fund itself such that all maturity mismatches are hedged, but this is impossible at thesystem level. Regulators are therefore trying to cope with the problem the other way around: forcing banksto build liquid reserves such that, while not matching outflows in terms of maturities, they ensure that ifa stress occurs then banks can withstand cash imbalances until the situation returns to normality.

6


2.1. The Regulatory Effort

In the regulator’s view the new requirements will have to be stringent: capital and liquidity resources willbe such that the financial system must have the strength to withstand a crisis of the size and persistencyof the recent one, without public support2.

Security will come at a cost.Several studies have tried to measure the cost for banks and the broader economyof the new rules as proposed by the Basel Committee in the December 2009 formulation. We will quotetwo authoritative ones:

• The Institute for International Finance3 has estimated that “the current calibration of regula-tory reform would subtract an annual average of about 0.6 percentage points from the pathof real GDP growth over the five year period 2011-15, and an average of about 0.3 percentagepoints from the growth path over the full ten year period, 2011-2020.The Euro Area would behit the hardest; Japan the least, with the United States somewhere in the middle” as per thefollowing table:

Cumulative Effects Results in Summary

difference between regulatory change and base scenario

Difference in average rates: 2011-15 2011-20

Real GDP growth difference

United States -.05 -.03

Euro Area -.09 -.05

Japan -.04 -.01

G3 (GDP-weighted) -.06 -.03

• According to a McKinsey survey on European banks4, the impact of the liquidity ratios in theircurrent versions is estimated as follows:

• LCR: increase in liquid asset holdings in the region of Eur. 2 trillion.

• NSFR: increase in long-term funding (>1 year) in the range of Eur.3.5 to 5.0 trillion (this comparesto current outstanding long-term unsecured debt of Eur. 10 trillion).

• As to banks’profitability,McKinsey provides an estimate that covers both the liquidity ratios and theproposed tougher requirements on capital.McKinsey expects that the return on equity (ROE) of thebanking sector in Europe could decline by 5% compared to its long-term average of 15%

At the end of the G20 meeting in Toronto, Jaime Caruana, General Manager of the Bank for InternationalSettlements, tried to reassure banks that the new rules “will not undermine economic growth”, and willonly have a “small and temporary” effect on demand5. Basel Committee representatives mentioned thatestimates such as those quoted above could be excessively pessimistic, particularly on the basis that theydo not sufficiently account for the dynamics of the financial system and the behavior of its operators.Investors are likely to require a lower return on capital as a result of the perception of a lower risk. Also,banks are likely to build new products designed to match the more stringent regulatory requirements interms of both capital and medium/long term funding, and to smoothen the impact of new regulation oncosts.The recent announcement by Unicredit, the third largest European bank by market value, of a newhybrid product designed to match the new capital eligibility requirements is an example in this direction.

2 See Mario Draghi, Chairman of the Financial Stability Board, Letter to the G20 Meeting in Toronto, July 20103 Institute for International Finance, Interim Report on the Cumulative Impact on the Global Economy of Proposed Changes in the Banking Regulatory

Framework, June 20104 Basel III: What the draft proposals might mean for European banking, McKinsey on Corporate & Investment Banking, Summer 20105 The Financial Times, Move to reassure banks on tough rules, 5 July 2010

7


Nevertheless, the regulatory effort is extremely delicate. The regulators are moving across unknownground and it is difficult to assess the potential outcome of the new constraints.The decisions stated in theBIS 26 July statement, which defined very long transition periods for the newest and most challengingsubset of the new planned requirements (the Leverage Ratio and the Net Stable Funding Ratio), are a clearsign of this. By delaying the new regulations until after a full economic cycle has taken place, supervisorswant to avoid them hitting economies while a difficult recovery is taking its course, while at the same timecreating the space to assess their potential impact on all the phases of the cycle.

2.2. The New Liquidity Requirements

BCBS’s proposals for enhanced liquidity requirements were presented in the consultative document:International Framework for Liquidity Risk Measurement,Standards and Monitoring,published in December2009, subsequently amended in the communiqué of 26 July, 2010 and finally formalized on 12 September.

The Committee proposes to introduce two new ratios (Liquidity Coverage Ratio and Net Stable FundingRatio) that banks must maintain as a minimum at all times to ensure they maintain sufficient liquidity towithstand cash obligations even under stress. For the NSFR, the 26 July statement set forth an “observationphase to address any unintended consequences across business models or funding structures”.The NSFRwill be finalized and introduced as a regulatory standard on 1 January 2018.

The ratios are as follows:

• Liquidity Coverage Ratio: focuses on the shorter end of the time horizon and is aimed at ensuringthat each bank owns liquid resources to such an amount that short-term cash obligations arefulfilled even under a severe stress.

The ratio requires banks hold enough liquid assets to offset the sum of all cash outflowsexpected over the next 30 days:

Stock of High-Quality Liquid Assets > = 100%

Net Cash Outflows over a 30-day Period

Liquid assets are considered in terms of market value,to which standardized haircuts are applied dependingon type of asset and grade of liquidity. Net cash outflows are calculated by aggregating the bank’s assetsand liabilities under standardized categories and applying to each of them standardized coefficientsreflecting predefined stress assumptions.For example,a 75% factor applied to unsecured wholesale fundingin the denominator of the ratio entails an assumption that 75% of the currently outstanding amount willrun off within the next 30 days.

• Net Stable Funding Ratio: looks at a medium-term horizon and focuses on the structuralbalance between maturities of a bank’s assets and liabilities. It is aimed at preventing banksfrom exposing themselves to extreme maturity transformation risks by funding medium andlong-term assets with very short-term liabilities. It was this practice that generated a massiverisk in 2007, which turned into a systemic liquidity shortage when major short-term liquiditychannels (especially those linked to securitized products) suddenly dried up.

8


The NSFR requires banks to have enough funding to last at least one year to compensate forall cash needs expected to occur beyond the same deadline:

Available Amount of Stable Funding >= 100%

Required Amount of Stable Funding

Available Amount of Stable Funding is made up by cash, equity and liabilities which areexpected to remain with the bank for at least one year, either because they have a longercontractual maturity,or because they can be considered “sticky”even if their contractual maturityfalls within that year (as is the case, for instance, for a share of retail deposits). Required Amountof Stable Funding is the amount of assets that are not expected to be reimbursed for at leastone year (and therefore need to be funded for at least this period) and cash outflows expectedto occur beyond one year as a result of contingent liabilities.

As well the LCR, the NSFR is calculated by aggregating a bank’s assets and liabilities (including contingentliabilities) into standardized categories and applying a set of coefficients reflecting standard scenarioassumptions regarding, for instance, the “stickiness” of the bank’s deposit base.

No favourable treatment is envisaged for the following instruments, for which banks must therefore have100% Stable Funding:

- Securitizable assets.

- Assets from securitizations (unless for covered bonds).

- Securities issued by banks or other financial institutions.

- Any security with rating lower than A-.

• Monitoring tools: in addition to the ratios, the Committee has listed a set of monitoring metricsthat should be considered as the minimum types of information supervisors should use in theirmonitoring activity.These include:

- Contractual maturity mismatch.

- Concentration of funding.

- Available unencumbered assets.

- Market-related monitoring tools: asset prices and liquidity, CDS spreads, equity prices, etc.

9


2.3. The New Liquidity Ratios: Tasks and Issues

2.3. 1. Insufficiency of Standardized Figures

The Basel proposals entail all banks maintaining the liquid asset buffer such that the Liquidity CoverageRatio is above 100% at all times. However, the Basel Committee has made clear that the ratio requirementsmust be seen in conjunction with the principles for liquidity risk management best practice set forth in the2008 document,6 and should by no means override these.

In other words, compliance with the ratios should be seen as a minimum requirement and should not betaken in itself as a sufficient indicator of soundness or stability.

In fact,exclusively looking at the two ratios could leave critical weaknesses and exposures completely hidden:

- The ratios only look at liquidity gaps in defined time horizons. No information is provided aboutliquidity exposures in other periods (from 30 days to one year and from one year onwards).A bank could have very substantial liquidity exposures, for instance on month two, and beperfectly compliant with the LCR.

- The ratios are calculated with pre-defined standard aggregations and stress assumptions(a one-size-fits-all approach).The significance of standardized aggregations and stress assump-tions can differ substantially across banks with different sizes and business models, thoseoperating in different countries, etc. For instance, a standardized assumption on the runoff ofdeposits in case of stress can be too loose in one situation and unnecessarily strict in another.

- The observation periods are standardized irrespective of individual banks’ business models.For instance, if a bank is heavily involved in correspondent banking, clearing and settlementactivities, then 30 days could be a very long-term horizon, as opposed to a bank heavily focusedon the retail deposit base, where 30 days would be considered a short-term observation period.

It is imperative that on top of complying with the regulatory ratios, each bank defines its own risk appetiteand runs internal stress tests for liquidity exposures that reflect its individual business model and vulner-abilities. Each bank should then define the required amount of the liquid asset buffer on this basis, i.e.independently of the regulatory requirements. In this context the Basel ratio should be seen as an externalminimum constraint,but the possibility that the optimal buffer is higher than that required by the regulatoryratios should not be ruled out.We will get back to this topic later in this document.

6 Basel Committee for Banking Supervision, Principles for sound liquidity risk management and supervision, September 2008

10


2.3. 2. Building Differentiated Incentives to Traditional Banking vs. Speculative Trading

It is critical that the final calibration of requirements generates a grid of balanced incentives for differentkinds of banking activities. Retail and corporate deposit taking and lending should be granted a favourabletreatment as opposed to more speculative activities. A fundamental rationale for this is that a favourabletreatment should be reserved for those parts of banking that respond to a public interest at the broadereconomy level. Liquidity mismatch risk should be more politically and socially acceptable to the extentthat it responds to the crucial role of the banking industry of efficiently reallocating financial resourcesacross lending and borrowing sectors of the economy.

From a more technical perspective, customer deposits deserve a favourable treatment as they are themost stable funding source for a bank. During the crisis, banks with a large deposit base performed betterthan others.This is also consistent with the outcome of an Oliver Wyman survey:“Banks which had a solidfunding base (defined as the ratio between customer deposits, long-term debt and equity capital overliabilities) have performed significantly better on average compared to banks relying on shorter-termfunding options. In this sample of selected global banks in developed markets, 80% of banks that had a‘solid funding ratio’before the crisis that was above 0.65 outperformed the industry average after the crisis,where as all banks below 0.65 underperformed significantly”7 as also shown by the following chart:

7 State of the Financial Services Industry Report, Oliver Wyman 2009

400

200

0

-200

-400

-600

-800

-1000

-1200

0.25 0.50 0.75 1.00

0.65

Industry average

Average Solid Funding Ratio Before-Crisis (Jan ‘04-July ‘07)

Solid funding ratio Before-Crisis contributes to shareholder valuecreation After-Crisis Selected global banks in developed markets

After-Crisis SPI (Aug ‘07-Dec ‘08)

Source: Bloomberg, Datastream and Olive Wyman analysis

11


The task of providing differentiated incentives for “traditional” banking as opposed to more speculativefinancial activity is apparent in how the Committee has addressed the new requirements for capital andleverage. Here, more speculative instruments such as derivatives (especially if traded over-the-counter)are subject to much higher capital requirements. Also, in the current formulation of the leverage ratio avariety of derivatives are fully considered as assets to be accounted for against capital, and netting ofderivatives is in principle forbidden.

When we come to the liquidity ratios, however, such differentiation of incentives is less clear. Indeed,the BIS has shown it will to move in this direction in the 26 July document by defining a more favourabletreatment of deposits from retail and small to medium-size enterprises both in the LCR and the NSFR, andof mortgages in the NSFR. Nevertheless, we believe the incentive to “traditional” lending could be furtherenhanced. For instance, loans with maturity below one year severely impact the required amount ofmedium/long-term funding, as it is assumed that banks will be forced to roll over beyond the one-yearhorizon 85% of such loans if granted to retail clients, and 50% if granted to non-financial corporations.In our view, this penalizing assumption could be smoothed.

Also, conditions could be defined under which a more favourable treatment in the NSFR is allowedfor medium/long-term maturity assets that are bound to be securitized in the short run. We will addresssecuritizations in a dedicated section later in this document.

2.3.3. Accounting for Bank’s Size

In their current formulation,the Basel ratios are uniform irrespective of the size of the banks to which they apply.

Even if two banks show equal ratios, the potential systemic impact of a liquidity issue can be totally differentdepending on the absolute amounts of their exposures. In this view, a uniform measure can prove unnec-essarily strict for smaller banks, while not providing the desired degree of protection against systemiceffects for larger banks.

We would therefore find it suitable that the individual bank’s size be taken into account in the definitionof the standard requirement.

2.3.4. Need to Raise New Capital

One of the main issues with the new regulatory requirements is that banks may prove unable to raise capitalor medium/long-term funding in the amounts required.

It should be noted that there is a clear interdependence between capital and liquidity requirements. In fact:

• In the NSFR, capital is directly and fully eligible as a “stable” funding source in support ofmedium/long term liquidity needs.

• In the LCR,capital is only indirectly considered as an eligible source of liquidity for the calculationof the ratio: it is taken into account only to the extent it is invested into eligible liquid assets.

Indeed, the Committee’s idea is clearly for banks first to raise the amounts of new capital as required, andthen invest it to build the liquid asset buffer.

However, the implication of this is that investors would be asked for new capital under the certainty thatit will be invested into low-yield instruments. Investors might be ready to accept a lower return on theircapital in view of lower risk, but this will need to be tested.

12


If banks are not able to raise new capital in the amounts required then the only alternative would be toreduce assets.This could hit the broader economy by capping the amount of credit available to it.

This undesired impact would be exacerbated in the context of a constrained leverage ratio, whereby liquidassets are fully considered in the calculation of total assets to be put against capital as per the currentproposal from the Basel Committee. Indeed, this provision does not appear to be fully justified in theoryand could in our view be lifted.This would allow banks using borrowed resources to fund the liquid assetbuffer, and would therefore grant them a greater degree of freedom in making strategic decisions aboutthe amount of credit available for clients.

2.3.5. Securitization Disruption

In its current formulation, the NSFR has the potential to disrupt the market for asset securitizations. Indeed,the core attractiveness of securitization is its capacity to transform assets that are illiquid in nature intoliquid instruments that can be managed and traded on a short-term basis. In addition, securitizations caneffectively provide reserve liquidity to the extent that asset-backed securities (ABS) are eligible as collateralfor repos with the relevant central bank (as is currently the case under certain conditions with theEuropean Central Bank).This characteristic is denied by the NSFR from two perspectives:

- Perspective of a bank willing to invest in asset-backed securities: ABS are not eligible asliquid assets to any extent. All holdings of ABS with a maturity exceeding one year are 100%accounted for in the determination of required stable funding and must be matched withmedium/long-term funding.

- Perspective of a bank willing to grant medium/long-term credit to its customers in the form ofmortgages,credit card loans,personal loans etc: the NSFR states that loans with maturity exceedingone year must be funded with medium/long-term finance up to percentages that depend onthe loan credit quality and are completely independent of the possibility of being securitized.As a result, lending banks cannot draw any benefits in terms of treasury from securitizations.

The misuse of ABS was indeed a main instigating factor for the crisis in 2007. However, the case forsecuritization is still valid, and it is important to avoid throwing out the baby with the bathwater.The NSFRapproach appears unnecessarily strict. ABS misuse should be addressed by specific regulation.The NSFRshould instead recognize the case for securitization and allow banks to manage liquidity exposuresaccordingly. Examples of possible approaches are as follows:

- Investor’s perspective: grant a more favorable treatment to ABS by applying coefficients thatdo not imply full medium/long-term funding.

- Lender’s perspective: allow banks with an established and demonstrable record of assetsecuritization to segregate certain loans that have been issued in view of being securitizedover a specified time horizon and fund them over that specified time horizon; allow ABS that areeligible for central bank repo to be accounted for as liquid assets.

13


2.3.6. Raising New Medium/Long-Term Finance (NSFR)

Even after considering new liquid resources incoming from capital increases, banks would probably stillneed to raise new medium/long-term funding in quite substantial amounts to comply with the NSFR.It should be noted that the NSFR provides a clear counter-incentive for banks to invest in securities issuedby other banks as these, no matter how actively traded, cannot be considered as liquid assets and cannotcount 100% as required stable funding.Therefore, banks should seek new debt from non-financial sectors.

It might not be obvious that non-financial investors are ready to provide medium/long-term finance inthe amounts required. And even if they were then the question could be asked, at what prices. Again, whatis at stake here is the impact on the broader economy.To the extent that banks succeed in raising debt inthe required amount, they would probably try to pass additional funding costs on to customers such thatthe cost of credit would increase. To the extent that sufficient debt is not available, the only option forbanks would be to reduce the amount of medium/long-term credit available.

Indeed, concerns about the possible impacts of NSFR on banks and the economy are the core reason fordelaying the introduction of the NSFR as a requirement until 2018. We believe the NSFR addresses a realneed in the banking industry because it sets a limit on the ability to create maturity mismatches in theshort run, which is important from a systemic perspective9. It should also be noted that the NSFR does notprevent banks from assuming maturity mismatches, as they could still fund 30-year loans with 12-month-plus-one-day funding. What the NSFR is addressing is the building of massive maturity mismatches overthe short run, such that not enough time would be left to find viable solutions in case of generalized andpersistent stress.

At the same time, we would favor smoothing the calculation of the ratio in a number of ways. For instance,the observation period could be shortened to 6 months. While substantially reducing the impact onborrowing costs for banks,such a time horizon would probably still provide a sufficient timeframe for findingsolutions in case of systemic and persistent stress. At the same time, it would generate a strong incentiveto increase the average maturity of interbank deposits, which is currently unnaturally and undesirablystuck to the shortest end of the maturity ladder.

Also, a looser approach to ABS and securitizations as suggested in Paragraph 3.2.5 above would help inthis respect.

9 This type of requirement is not new in liquidity risk supervision.The Bank of Italy required for some years local banks to comply with the so-called “MaturityTransformation Rule”, that implied a limitation on the possibility to get exposed to maturity mismatches.This rule was lifted some years ago, well before thestart of the crisis.

14


2.3.7. Distorting Bond Markets

The LCR is likely to generate a huge shift in demand towards assets eligible for inclusion in the liquid assetbuffer – in essence, G8 liquid government bonds. This would act as a de facto constraint for banks thatfinance government deficits. Given the current imbalanced situation of public finances worldwide, thismight be a desirable effect that would help preserve financial stability at the global level. But at the sametime this would have undesirable consequences:

• Bond markets would be distorted: liquidity and prices of liquid assets would be artificiallyincreased and their yield depressed, while the market for instruments not eligible for the bufferwould be negatively impacted, with reduced liquidity, higher yields and lower prices.

• The above would negatively impact the ability of non-banking industries to raise funds throughbond markets.

• Market prices would no longer be indicative of the market’s risk appetite and requiredrisk/reward profiles. This would apply to both instruments eligible and non-eligible as liquidassets, for the reasons explained above.

The BIS 26 July statement has broadened the acceptable criteria for security that are eligible as liquidassets.The potential for the above impacts is therefore reduced, although not in any way removed. It willbe important that distortions are closely monitored and assessed, such that further calibration can bedone to minimize such effects.

2.3.8. Reshaping Interbank Deposit Markets

A clear task of the Basel Committee is to reduce individual bank’s dependence on interbank funding.The ratios have been built on the assumption that from a system point of view, interbank financing is onlyapparent and does not provide any safety if there is a systemic funding liquidity issue.

As a result, neither in the LCR nor in the NSFR is interbank funding granted any favorable treatment as aneligible source of liquidity in relation to potential cash obligations. Both ratios are calculated upon thestress assumption that no maturing interbank liabilities will be rolled over and no new interbank fundingwill be available.

In addition, if a bank holds securities issued by banks or other financial institutions, no matter how liquid,in NSFR these have a 100% weighting in the calculation of Required Stable Funding. As a result, theycannot be treated as liquid assets and are fully considered in determining the minimum required amountfor medium/long term funding.

Therefore, banks will be allowed to rely on outstanding interbank funding only to the extent they havea contractually formalized right to avoid repayment for more than 30 days (LCR) or for more than oneyear (NSFR).

While in agreement with this approach, we nevertheless believe that the requirement might be smoothedin view of encouraging a reshaping of the interbank deposit market towards a more desirable concentrationof trades on longer maturities,such as six months and above,compared to the current one month and below.

15


2.4. Survival Horizon Models: Optimizing the Liquidity Buffer in a Comprehensive Risk Management Framework

Despite ‘survival horizon’, or ‘survival period’, models not being expressly included by the Basel Committeein its list of mandatory metrics banks are required to use in liquidity risk assessment, they are gaining spaceamong supervisors for their ability to provide a synthetic indicator of a bank’s resilience against liquiditystresses. The LCR is in fact based on a survival horizon model with a 30-day observation period andstandardized stress assumptions. Australia’s regulator, APRA, in 2009 released a consultation paper where itenvisaged a new regulatory regime for liquidity risk, including an obligation for banks to provide survivalhorizon analysis10. The UK’s regulator, the FSA, while not expressly imposing SH-modeled reporting, diddisclose during public hearings that it will use banks’ reported figures to feed SH models in order to assesstheir resilience against stress.

Survival horizon models are based on a comparison of ‘forward liquidity exposure’, i.e. the sum of expectedcash flows over a defined period, to the amount of cash that the bank can expect to raise by selling orpledging its liquefiable assets over the same period.Both terms of the comparison are made subject to stressassumptions of defined severity.The survival horizon is the period over which the existing liquid or liquefiableresources are sufficient to support all expected cash outflows under the defined stress assumptions.

Survival horizon models are effective in delivering a synthetic indicator of a bank’s resistance to stress,after integrating the potential impact of different kinds of liquidity risk (e.g. market and funding liquidityrisk) and a variety of scenario assumptions. As such, they lend themselves to providing a single measurethat can be used as a reference for the definition of the bank’s risk tolerance.

2.4.1. The Relationship between the Basel Ratios and Bank – Specific Survival Horizon Models

The new liquidity ratios will have a key role in defining banks’ strategies:

- The Liquidity Coverage Ratio will determine the minimum amount of liquid assets givenexpected cash flows over the short run (unless the results of bank-specific stress tests mandateholding greater amounts of liquid assets).

- The Net Stable Funding Ratio will determine the minimum amount of medium/long term fundinggiven capital, medium/long lending business and contingent liabilities.

The need to hold substantially greater amounts of liquid assets will oblige banks to switch part of theirinvestments from more profitable assets to high-quality, typically low-yield instruments. This trade-offmight be further emphasized by the envisaged leverage ratio: in the current proposal, liquid asset holdingsare fully taken into account in the calculation of the leverage ratio – although the Basel Committee hasstated it might consider, after proper impact assessment, excluding certain categories of liquid assets fromthe calculation of the leverage ratio.

However, Basel Committee documents make it clear that the regulatory ratios should not be seen asabsolute. Individual requirements for liquid asset holdings must be defined after the results of individualstress tests, and the regulatory ratios should rather be seen as a regulatory minimum.

10 APRA’s prudential approach to ADI liquidity risk, Discussion Paper, 11 September 2009

16


Therefore, the required amount of liquid assets should be defined as the maximum of the results of theregulatory ratios and the amount resulting from individual stress testing.

Bank-specific stress tests should be based on assumptions designed to match the bank’s specific context,business model and vulnerabilities.As such,stress tests should tend to stress business strategies and businessmodels rather than external risk factors.We would mention reverse stress testing as a particularly suitabletechnique for this task.

2.4.2. Calculating the Amount of the Liquidity Buffer

The Basel Committee has outlined that banks’ stress tests must not be seen as an isolated exercise butrather be seamlessly inserted into their organizational processes and provide inputs for decision-makingand action.

The required liquid asset buffer should be defined as the minimum amount that ensures the bank’s abilityto fulfil expected cash obligations. It therefore has critical implications in terms of allocation of resourcesand risk-taking strategy. As a result, its calculation should be performed within the framework of a processthat has its core in the bank’s Risk Tolerance Policy and includes procedures for monitoring, reportingand decision-making. In our view, such a process would be most effectively supported by a stress-basedsurvival horizon model.

At a high level, the main steps of the process can be schematized as follows:

i) The bank’s board formally defines the bank’s liquidity risk tolerance, i.e. the maximum amountof risk it is willing to bear in its activity. This maximum risk should be defined under a stressscenario and expressed in terms of indicators that can be continuously monitored, controlledand reported. A ‘risk tolerance policy’ document should include, among other things, the following key elements:

• A synthetic indicator of the bank’s risk appetite, that the bank will continuouslymonitor and control; and

• The level of severity of the stress assumptions, as well as the type(s) of scenario (i.e.idiosyncratic,market-wide,combined) to be used in monitoring the bank’s compliancewith the defined risk threshold.

In our view, a survival horizon model is a particularly suitable framework for defining thesynthetic indicator of liquidity risk appetite. As an example, the liquidity risk appetite couldbe synthetically expressed as the capability to survive for two weeks under a high-severitystress scenario and for three months under a mild-severity stress scenario.

ii) Based upon indicators from the risk tolerance policy, the bank defines in detail the stress scenario assumptions that it will periodically test for its survivability capacity.

iii) The bank calculates its survival horizon under the defined stress assumptions. The required amount of liquid assets is the one that ensures matching the minimum survival period as defined in the risk tolerance policy.

iv) The bank compares the minimum liquid asset buffer resulting from the above process to the regulatory amount based on the Basel ratio. The greatest of the two is the required liquid asset buffer.

17


The above schematic only refers to a fraction of the overall process. Indeed,both the Basel Committee and theCommittee of European Banking Supervisors (CEBS) guidance documents make clear that internal processesneed to ensure full consistency and integration of risk tolerance definition, stress testing, risk monitoringand control, and decision making to prevent risk from climbing above the defined risk appetite thresholds.

The following summary figure provides a schematic:

2.4.3. Optimising the Liquid Asset Buffer

The liquid asset buffer is typically composed of high-quality, low-yield instruments. Banks will thereforeneed to optimize the amount of liquid assets by actively managing the portfolio so that it is kept as closeto the minimum required amount as possible.

In this perspective, the following implications of optimizing the liquid asset buffer should be considered:

- The CEBS clarifies in its guidance document11 that the global amount of the liquid asset buffershould be entirely driven by the longer end of the stress scenario horizon.However, results on theshorter end should be relevant for defining the composition of the buffer, as only instrumentsthat can be very quickly liquefied should be held as a shield against short-term, unpredictedsevere scenarios.

- Actively managing the liquid asset portfolio implies that information relevant for measuringthe required minimum buffer is available as frequently as possible. In other words, the bankshould be able to obtain up-to-date reports by re-running cash flow projections and liquidasset price simulations under the defined stress assumptions with a high frequency in order tohave the possibility of adjusting downwards the required amount of the buffer.

- Also, frequently updating projections and stressed simulations should be seen as a prerequisiteto keeping the portfolio at a level which is very close to the required minimum. High frequencyupdates of current and simulated data would in fact ensure that the bank is prompt in catchingearly-warning signals that would trigger an increase of the buffer amount. If simulated data areto remain static for long intervals due to an inability to update the stressed simulations, thebuffer should be prudentially maintained above the minimum.

- The stress assumptions used to build the model underlying the required buffer amount shouldalso be reviewed on a regular basis.

- Cash flow projections should be available over the time periods and with the time bucketgranularity required by the model chosen for defining the risk tolerance. Only looking atcumulative cash inflows and outflows over a defined time period will not ensure protectionagainst possible stress, as cash outflow needs to be fulfilled whenever it occurs, such that theadequacy of the buffer must be assessed for each day in the observation period.

11 Committee of European Banking Supervisors, Guidelines on Liquidity Buffers & Survival Periods, 9 December 2009

Risk Tolerance Policy Stress Test

LimitsEarly Warnings Contingency Funding Plans

Liquid Asset Buffer Survival Horizon

18


Example: Assume a bank’s risk tolerance policy sets the survival horizon at one month.To assesswhether the liquid asset buffer is sufficient for this task, the bank compares the current bufferto the net of cumulative cash inflows and outflows over 30 days under the internally definedstress assumptions. Assume however that a big cash outflow is expected on day five, and anequally big cash inflow is expected on day 29. In the cumulative 30-day cash flow projectionthe two flows will offset each other so that the spike in cash needs on day five will remain hidden.As a result, the buffer may prove insufficient to cover the cash requirement on days five to 29.

3. Proposals Regarding CapitalBasel II regulations that were finalized in 2004 were based on two key assumptions: the overall level ofcapital in the system is sufficient, but there is a need to increase the risk sensitivity within the frameworkto promote better risk management and to reduce regulatory arbitrage. Therefore the focus of Basel II,Pillar 1 was in the definition of risk-weighted assets. Regulators did recognize the need to refine thedefinition of the capital components, but agreed to revisit the issue after ratification of Basel II.The financialcrisis expedited the need for re-definition of capital since items that were considered Tier 1 capital couldnot absorb losses as a going concern.The predominant form of Tier 1 capital must now be common sharesand retained earnings, and the remaining part must be comprised of instruments that are subordinated,have fully discretionary non-cumulative dividends or coupons,and have neither a maturity nor an incentiveto redeem. Also, the risk sensitivity assumptions underlying various transaction types, especiallysecuritizations and derivatives, was found to be insufficient during the financial crisis. With theDecember 2009 paper on Strengthening the Resilience of the Banking System, the Basel Committeeshowed that it intends to increase the quality, quantity and international consistency of the capital base,while also increasing capital requirements for certain types of transactions and obligors. In addition,regulators are also introducing a non-risk based leverage ratio to reduce build up of leverage in the overallsystem.BCBS is also introducing rules to reduce pro-cyclicality inherent in the Basel II framework.This sectionsummarizes the salient features in the December 2009 proposals as well as subsequent communiquésrelated to capital base, risk coverage, leverage ratio, countercyclical buffers and systemic risk, and highlightssome of the potential inconsistencies and perverse incentives inherent in the rules as currently proposed.

3.1. Capital Base

The proposal addresses the shortcomings in the current capital framework in terms of the quality of thecapital, application of regulatory adjustments and the lack of international harmonisation in the way theregulatory adjustments are calculated.The main drawback in the current framework is that the regulatoryadjustments are applied either to a combination of Tier 1 and Tier 2 capital or only to Tier 1 capital, asopposed to the common equity component, which provides the real loss absorbing capacity as a goingconcern. Under the current proposals, Tier 1 capital is defined as items that can absorb losses under agoing concern assumption,and Tier 2 is defined as capital that can be used to offset losses as a gone concern.Tier 3 capital, which was allowed to offset market risk under Basel II, will be completely eliminated.

The qualifying criteria for the classification of the capital instruments into common equity,Tier 1 and Tier 2capital have all been made more stringent.When the draft proposal was introduced in December 2009,the Basel Committee had introduced stringent measures on eligibility of minority interest for inclusion inthe common equity component of Tier 1 and deduction of deferred tax assets from Tier 1. In the July 2010communiqué, which outlines the broad agreement reached between the Board of Governors of the BaselCommittee, some of the earlier proposals have been softened. For example, the prudent recognition ofminority interest for a banking subsidiary is now allowed.

19


With a view to improving the transparency of the capital base, financial institutions would be required todisclose all components of the capital along with the regulatory adjustments. In addition, banks arerequired to provide a reconciliation of the regulatory capital elements back to the audited financial statements.

A stricter definition of capital will make it more expensive as well as be a binding constraint in businessdecision making.

3.2. Risk Coverage

In addition to tightening the definition of capital, the new rules also propose to increase capital require-ments for certain types of transactions and obligors.There are multiple areas in which the regulators wantto increase capital requirements. A major thrust of the proposed Basel III rules focus on counterpartycredit risk (CCR) arising from bank’s derivative, repo and securities financing transactions (SFT) operations.The limitations of the current practices were highlighted during the financial crisis, especially by theLehman collapse. BCBS has focused heavily on this section, with detailed guidance on addressing generaland specific wrong-way risks,accounting for CVA losses with capital,requiring higher risk weights for financialcounterparties, providing incentives for banks to move trades to central counterparties, strengtheningthe collateral management function, increasing margin periods of risk, as well as stress testing and backtesting requirements.

3.2.1. Addressing General Wrong-way Risk – Stressed EEPE

General wrong-way risk occurs when the creditworthiness of the counterparties are positively correlatedwith general market risk factors. During the financial crisis it was observed that the exposures to counter-parties increased precisely when their creditworthiness deteriorated.

To address this issue, BCBS requires banks calculate CCR using stressed parameters. Effective expectedpositive exposures (EEPE) is calculated by first calculating exposures under multiple (Monte Carlo)scenarios and time paths, and then calculating the average exposure taking into account roll-off effects.Under the current proposals,EEPE needs to be calculated using a three-year period that includes a one-yearstress period. Stressed EEPE needs to be used in regulatory capital calculations in case it exceeds the EEPEcalculated using current period market data.This requires the bank calculate two EEPEs and compare theresults on a periodic basis. Flexibility and performance of existing risk systems is critical to achieving thisrequirement without major system re-engineering.

The alpha add-on factor under Basel II already captures the general wrong-way risk. Therefore, usingstressed EEPE will double-count general wrong-way risk, which can have a substantial effect in a tradingbook that’s already encumbered with additional market risk and incremental risk charge (IRC) capitalrequirements. BCBS studies have shown that own estimates of alpha (ratio of bank internal estimate ofeconomic capital based on stochastic exposures to economic capital based on EPE) are subject tosignificant variations across banks due to mis-specifications of the models, particularly for exposures withnon-linear risk profiles.The committee intends to strengthen the requirements for own estimates of alphato address this issue.There is industry demand to allow alpha based on counterparty, industry or productcharacteristics and not a single alpha across all counterparties.For banks that can implement such a system,alpha will provide a more useful measure of general wrong-way risk than the proposed stressed EPE.After the quantitative impact study, if it appears that more capital is required to support counterpartytrading activity then it is more desirable to apply a transparent scalar to increase capital to the desiredlevel, instead of double counting through alpha and stressed EPE.

20


3.2.2. Capturing CVA Losses

During the financial crisis capital charge for mark-to-market losses was greater than the losses due todefaults. Since global banks have diverging practices for CVA calculations, BCBS is introducing a simplified“bond equivalent of the counterparty exposure” approach to calculating CVA risk. This entails modellingall of the counterparty’s exposures as a zero coupon bond with a notional amount equal to EEPE anda maturity equal to effective maturity (M) of the exposure profile. CVA risk is defined as the regulatorymarket risk charge for this stylized position calculated using a 1-year horizon instead of the 10-day marketrisk horizon, excluding the incremental risk charge.

There is a double counting issue here as well.The existing maturity adjustment in the risk-weighted assets(RWA) formulas already account for migration risk with an analytical approximation.This maturity adjustmentshould be removed if the CVA charge is included in capital to avoid double counting.

The bond equivalent approach is introduced to achieve methodological consistency across banks as wellas to provide a simple, intuitive approach the Committee assumes will reduce the system and method-ological burden for banks. However, banks that already calculate EPE have the full distribution of potentialfuture exposure (PFE) profiles at their disposal since EPE is essentially a function of the PFE distribution.Therefore, it is an extra burden for banks that already calculate EPE using a simulation approach.

Furthermore, the “simple” bond equivalent approach can in fact end up causing perverse incentives.Depending on the shape of the PFE profile using only EPE and M to represent the full profile can resultin over- or under-estimation of the sensitivities and the required hedges. For example, in the case of adownward sloping exposure profile, EPE will be close to the time zero exposure (which is the maximumexposure when the exposure is decreasing over time). In a downward sloping profile, the bank will needlarger short-term hedges and smaller longer term hedges to effectively hedge counterparty risk that isdecreasing over time. However, the bond equivalent approach, which assumes a single maturity date (setto effective maturity) and a constant exposure, will require a hedge that is equal to EEPE for duration equalto effective maturity.This will result in an inconsistency between the hedges required for the actual exposureversus the hedge required to minimize regulatory capital. Therefore, the bond equivalent approach is apoor approximation to CVA risk and does not properly capture hedge effectiveness. From a sensitivityand stress testing perspective, using the bond equivalent approach to model CVA risk will result in largersensitivities for maturities less than M, and no sensitivity for maturities longer than M. Given that full PFEprofile and hedge information is already available as inputs to the EPE calculation, it would be much moreproductive to allow banks to use internal models for capturing CVA leveraging full PFE profile.

Furthermore, if the bond equivalent approach is ratified, banks will have to calculate CVA twice – once forinternal purposes and then for regulatory purposes.This will overly burden banks to maintain two systemsin addition to being contrary to the spirit of “use tests”.

The bond equivalent method, however, does provide an opportunity for banks that do not have a CVAsystem to calculate CVA in a simplified manner. It is desirable to provide incentives for banks to developfully-fledged CVA systems and use it for regulatory capital purposes.This will be consistent with the currentBasel principles of mandating a simplified approach while providing banks with the incentive to move tointernal models subject to supervisory review.

21


BCBS has realized some of the shortcomings of the bond equivalent approach and is addressing some ofthe issues.The July 2010 communiqué was encouraging as it will allow the bond equivalent approach toaddress hedging, risk capture, effective maturity and double counting, although it is lacking in detail.It is encouraging that BCBS has mentioned addressing double counting, as we encounter that in manyplaces in the document and this needs to be revised to ensure the rules are internally consistent. BCBShas indicated that it will undertake a more fundamental review of the trading book and look at moreadvanced alternatives to the bond equivalent approach.We hope this will lead to full approval of internalCVA methodology and systems.

3.2.3. Specific Wrong-way Risks

Counterparty risk management standards are being raised where there is specific wrong way risk i.e. wherecounterparty exposures increase when the credit quality of the counterparty deteriorates. Banks are nowrequired to perform stress testing and scenario analyses to identify risk factors that are positively correlatedwith counterparty credit worthiness, and these scenarios should address the possibility of severe shocksoccurring when relationships between risk factors change.

Banks should manage wrong-way risk by product, region and industry or by other relevant categories.Transactions where specific wrong way risk (future exposure to a specific counterparty is highly correlatedwith the counterparty’s creditworthiness) has been identified will require significantly higher exposuremeasures, which in turn results in higher capital charges:

• For single name credit default swaps (CDS) with specific wrong way risk (where the single nameand issuer have a legal connection) exposure at default (EAD) = Notional Amount.

• For equity derivatives referencing a single company (with specific wrong-way risk): EAD = valueof the derivative under assumption of default of the underlying.

This is in contrast to the Basel II treatment of calculating EAD under the ‘current exposure method’ forcredit derivatives and equity derivatives as mark-to-market plus an add-on. Add-on factors for credit deriv-atives are 5% for a qualifying reference obligation and 10% for non-qualifying. The new treatment canincrease EAD and capital easily by 10 times or more for derivatives where specific wrong-way risk existsunder the current exposure method.The impact under the internal model method (IMM) can be possiblyhigher. This provides a strong disincentive to trade such contracts outside of central counterparties.

3.2.4. Higher Risk Weights for Financial Institutions

Empirical studies by the Basel Committee have indicated that asset value correlations for financial firmsare, in relative terms, 25% or higher than those of non-financial firms. A multiplicative factor of 1.25 (to becalibrated after a quantitative impact study) is to be applied on the formula used to compute the corre-lation for exposures to financial intermediaries that are regulated banks, broker/dealers and insurancecompanies with assets of over $100 billion as well as other (unregulated) financial intermediaries, such ashedge funds/financial guarantors.

This clause can increase capital to low probability of default (PD),high asset value correlation (AVC) financialinstitutions (typical profile being large inter-connected financial institution) by approximately 35% dueto the non-linear relationship between capital and AVC.

22


In the Basel II internal ratings-based (IRB) formulas,AVC is used to capture the default risk part of capital,andthe ‘maturity adjustment’ captures the migration risk. The justification given by BCBS is that “...financialinstitutions credit quality deteriorated in a highly correlated manner...”. If the intention is to capturemigration risk, it is best captured either by a modified maturity adjustment or through the CVA.To increasethe correlation value increases the default losses and not the migration losses.This is another instance ofpotential inconsistency in the proposals.

3.2.5. Increase Margin Period of Risk

The financial crisis has shown that the mandated margin periods of risk for regulatory capital calculationsunder-estimated the realized risk during the financial crisis. For transactions subject to daily re-marginingand mark-to-market valuation, the supervisory floor is set at five business days for netting sets consistingonly of repo-style transactions, and 10 business days for all other netting sets for calculating EAD withmargin agreements. A higher supervisory floor applies to all netting sets where the number of tradesexceeds 5,000, and for netting sets that contain one or more trades involving collateral that is illiquid, oran OTC derivative that cannot easily be replaced. If a netting set has experienced more than two margincall disputes over the previous two quarters then the margin period should be twice the supervisory floorfor that netting set. Proposed rules will substantially reduce the effect of netting and collateral on exposureand capital for repo,SFT and OTC derivatives, further providing incentives to move to central counterparties.

While all these rules are in the right direction and provide for a conservative approach, some of thethresholds are arbitrary. For example, the 5000 limit for netting sets is better considered guidance,and adjusted based on the capabilities of the collateral management system and the liquidity of theinstruments in the netting set.

3.2.6. Preclude Downgrade Triggers

Downgrade triggers are a popular credit mitigation technique banks use to cut off further lending tocounterparties when their ratings fall below a threshold unless they post extra collateral. During thefinancial crisis, however, some fallen angels downgraded so fast that the banks were not able to imposedowngrade triggers, and these clauses did not provide the expected credit mitigation. Therefore, thenew proposals require that banks using internal models do not take into account clauses in the collateralagreement that require receipt of collateral when credit quality deteriorates.This is a conservative practicein line with the spirit of the new proposals, since the likelihood of counterparties posting collateraldecreases when their rating goes down, especially during a crisis when the downgrade happens rapidly.

However, banks are required to take into account downgrade triggers imposed on them by their lenders.Going by the same logic, the bank will not be able to post collateral in a timely manner if its credit qualitydeteriorates rapidly. In order to make the rules symmetrical and internally consistent, it would make sensenot to model own bank downgrade triggers as well.

23


3.2.7. Collateral Management

BCBS intends to strengthen the standards for collateral management and initial margining under Pillar 2.BCBS supports the creation of a collateral management unit responsible for calculating and makingmargin calls and managing margin call disputes. On a daily basis, the collateral management unit wouldaccurately report levels of independent amounts, initial margins and variation margins. It would track theextent of reuse of collateral and the concentration to individual collateral asset classes. Reliable data oncollateral will enable the bank to use this data in PFE and EPE calculations.

In the Basel II framework, the standardized haircuts currently treat corporate debt and securitizations in thesame manner. During the crisis, the securitizations exhibited much higher price volatility than similarlyrated corporate debt. Therefore, collateral haircuts for securitization exposures are doubled relative tosimilar rated corporate debt. Further, BCBS has made re-securitizations ineligible as collateral goingforward, and stipulated strict controls around re-use of collateral.

There should be built-in regulatory incentives for banks to improve risk management systems.For example,for a bank that shows the regulator it has a comprehensive collateral management system and can easilyhandle large netting sets, the increase in margin risk period should not be applied when the netting setexceeds 5000.

Bank risk systems should be flexible enough to calculate exposure and capital under multiple assumptionssuch as different margin periods of risk, different initial and variation margins, so the sensitivity of thesespecific rules on the level of capital can be compared.

3.2.8. Central Counterparties

At the height of the financial crisis in 2008, major financial institutions could not tally exposures againstcounterparties across various transactions. Therefore, regulators and economic policy makers could notmake the right decisions in a timely manner. For example, when the US Treasury allowed Lehman Brothersto default it did not have a detailed picture of all the counterparties exposed to a Lehman default. If thisinformation was readily available the final outcome could have been different.

Exposures to the central counterparty will receive a near-zero risk weight (1-3%), providing banks with astrong incentive to move trades to the central counterparty clearing house (CCP). Given their systemicimportance, CCPs should be strictly supervised through rigorous standards, as a failure of a central coun-terparty could make the problem much worse. The upshot of these rules will be more standardized OTCcontracts clearing through CCPs.

3.2.9 Stressed PDs for Highly Leveraged Counterparties

New rules stipulate that PD for a highly levered counterparty should be estimated based on a period ofstressed volatilities.This again smacks of double counting. Leverage should be one of the factors alreadyconsidered, especially in rating financial counterparties. Furthermore, through-the-cycle (TTC) modelscapture stress periods in the rating estimate for all counterparties.

A better approach is for the committee to require banks to explicitly use leverage as a factor in determiningPDs for all counterparties.This would result in a more consistent treatment across all counterparties.

24


3.2.10. Stress Testing

Banks are required to have a comprehensive stress testing program for counterparty credit risk. Thisincludes ensuring complete trade capture and exposure aggregation across all forms of counterpartycredit risk in a timely manner so as to conduct regular stress tests. For all counterparties, banks shouldstress test principal market risk factors (e.g., interest rates, FX, equities, credit spreads and commodityprices) in order to identify outsized concentrations to specific directional sensitivities. Banks should alsoapply multi-factor stress tests which should analyze the impact of the portfolio under scenarios that reflectsevere economic and market events that occurred during the financial crisis, where broad market liquiditydecreased significantly and liquidating positions of a large financial intermediary created a large market impact.

Banks should also conduct reverse stress test to identify extreme but plausible scenarios that could resultin significant adverse outcomes on exposures and capital. This requires an integrated data and analyticplatform across all significant risk types.

The proposals stipulate periodicity of stress tests, and this seems overly prescriptive. Stress testing shouldbe an integral part of the bank risk management policy. However, stipulating periodicity and nature ofstress tests with a one-size fits all approach is not ideal. Larger, interconnected firms should have a morerobust and automated system for stress testing wrong-way risks and generating reverse stress tests.Bank systems should be able to adapt to have more frequent stress tests and the ability to perform ad-hocstress tests, especially during a crisis period.This type of requirement will be overly burdensome for smallerinstitutions with limited resources, and may also not be required since these institutions are typically notsystemically important.

3.2.11. Back Testing

Banks are required to conduct a regular back testing program, which would compare risk measuresgenerated by the model against realized outcomes. These requirements call for manipulation of largedatasets. A comprehensive back testing program calls for a system that can handle large batch systems aswell as provide the ability to set up ad-hoc queries.

3.2.12. Reduce Reliance on External Ratings

A major consequence under Basel II was to rely excessively on external ratings for regulatory capitalrequirements.This resulted in the neglect of bank’s own independent internal assessment of risks to a certaindegree. Ratings agencies have an incentive to produce “good ratings” since issuers, originators andinvestors all prefer “good ratings”. Given the Basel II rules, banks have an incentive to seek ratings just abovethe “cliff”. For example, the ‘standardized’approach prescribes a higher risk weight to corporate exposuresthat are rated below BB- (150%) than for unrated exposures (100%).This provides banks with an incentivenot to get ratings for companies that are likely to be rated below BB-.

With the new proposals low quality ratings would apply to unrated exposures that are pari passu orsubordinated to the low quality rating. Banks should internally assess if the risk weights applied (underthe standardized approach) are appropriate for their inherent risk. If it turns out that the inherent risk ishigher, then the bank should consider the higher degree of credit risk.BCBS also proposes the elimination ofthe A- minimum requirement for guarantors in the standardized approach and the foundation IRB approach.

25


3.3. Leverage Ratio

Another major cause for the financial crisis was the uncontrolled build up of leverage in the banking system.In order to constrain the build-up of leverage, the proposals reinforce the risk-based requirements with asimple non-risk-based “backstop”measure based on gross exposure.The July 2010 communiqué clarifiedthat the leverage ratio will be calculated after applying Basel II netting for all derivatives (including creditderivatives). In addition, a simple measure of potential future exposure based on the standardized factorsof the current exposure method (CEM) is to be applied to arrive at a “loan equivalent”amount for derivativeproducts.The leverage ratio would be calculated as an average over the quarter.

The 26 July document, confirmed by the 12 September press release, stated that the phases of thetransition to the adoption of the ratio are as such:

- Supervisory monitoring period from 2011 to 2012, focusing on developing templates to trackin a consistent manner the underlying components of the ratio.

- Parallel run period from 1 January 2013 to 1 January 2017, during which the leverage ratio andits components will be tracked. Bank disclosure of the leverage ratio and its components willstart on 1 January 2015.

- Migration to a Pillar 1 treatment from 1 January 2018 after proper calibration based on theresults of the parallel run.

The Committee proposes to test a minimum Tier 1 leverage ratio of 3% during the parallel run period. Forthe purposes of calibration, it is suggested that the new definition of Tier 1 capital as well as the ‘total capital’and ‘tangible common equity’ be used. The additional leverage ratio will act as a binding constraint forsome banks when deciding on whether to pursue new business. Therefore, banks will have to examinethe impact of a new transaction on both the capital ratio and leverage ratio. This will result in additionalbusiness and system impact.

There are substantial differences in accounting treatments among jurisdictions. As a result, the leverageratio will need to be calibrated thoroughly in order to avoid level playing field issues that could easily occur.

Accounting regimes lead to the largest variations. In particular, the use of International Financial ReportingStandards (IFRS) results in significantly higher total asset amounts, and therefore lower leverage ratios forsimilar exposures, than does the use of U.S. Generally Accepted Accounting Principles (GAAP).

The Committee agreed on the following design and calibration for the leverage ratio, which would serveas the basis for testing during the parallel run period:

a) For off-balance-sheet (OBS) items use uniform credit conversion factors (CCFs) with a 10% CCFfor unconditionally cancellable OBS commitments (subject to further review to ensure thatthe 10% CCF is appropriately conservative based on historical experience).

b) For all derivatives (including credit derivatives), apply Basel II netting plus a simple measure ofpotential future exposure based on the standardized factors of the current exposure method. Thisensures that all derivatives are converted in a consistent manner to a “loan equivalent”amount12.

12 Taken together, this approach would result in a strong treatment for OBS items. It would also strengthen the treatment of derivatives relative to the purelyaccounting based measure (and provide a simple way of addressing differences between IFRS and GAAP).

26


c) Since the leverage ratio proposed by the Basel Committee is non risk sensitive, it discriminatesagainst safer assets and liabilities. In fact less risky assets – which usually also have lowerreturns – will become less attractive, some of the most important examples being mortgagesand credit provision to small and medium-sized enterprises.

Banks that are facing the limits of the leverage ratio will have an incentive to take on riskier assets thatprovide higher returns, effectively increasing the risk appetite of the institution as a result of balance sheetconstraints that need to be taken into account alongside the risk management considerations. In particular,a differentiated adoption of Basel constraints across jurisdictions could prevent a competitive levelplaying field in terms of competition. Asset portfolios of banks that have fully adopted the Basel II frame-work (and therefore are using a risk sensitive approach for capital requirements purposes) are generallyless risky compared to the non Basel II compliant banks. Consequently, often the Basel II compliant bankshave a higher leverage ratio to compensate for the lower revenue that is a consequence of having saferassets on the balance sheet.These banks, in their attempt to optimize their balance sheet in terms of risk,will decrease the leverage ratio. But at the same time, in order to maintain the same expected returns, theywill have an incentive to invest in riskier assets. This is certainly an unwelcome outcome that the BaselCommittee should monitor carefully.

d) Moreover, the fact that no specific elements of different business models are taken intoaccount by developing only a single indicator is a major weakness in the concept.The leverageratio will impact the business model of the bank on an integral basis. As the other proposalsof the Basel Committee pertain to an array of different areas, the overlaps with the leverageratio will be numerous. In our opinion,the only way to properly assess the impact of the leverageratio is to see it in an integrated way with all the other proposals, including those regardingthe framework for liquidity risk. It is now clear that this compartmentalized approach, withadd-ons to meet minimum regulatory requirements, is seriously flawed.With the emergenceof Basel III we have an opportunity to reverse the trend towards splitting risk into differentsilos. However, we are concerned that the new rules could again fail to produce a frameworkthat refutes the compartmentalized way of thinking. This is seen in the way the BaselCommittee is addressing liquidity risk, and specifically in the lack of recognition of theconnections between leverage ratio, capital requirements and liquidity buffer.

e) Last, but not least there is a possibility that lending would shift from regulated banking to lessregulated financial institutions.The unwelcome result will likely be that in periods of financialdistress most of the risk that apparently will sit in these financial institutions will turn backto the banking organizations that were indirectly subsidizing the less regulated financialinstitution. What we have learnt during the recent crisis is that – and this is also importantfrom a reputational perspective – what ultimately matters is from where risk has originated,which of course will lead to the same answer: in the banking system.

27


3.4. Counter-Cyclical Capital Buffers

The recent financial crisis highlighted the pro-cyclical amplification of financial shocks. The measuresproposed in the consultative paper are designed to dampen excess cyclicality, promote forward lookingprovisioning, conserve capital for use in periods of stress and protect the overall banking system fromexcessive credit growth. The BCBS issued a consultative document regarding its proposal for a counter-cyclical capital buffer (the “Proposal”)13.

In this consultative paper, the BCBS stated that the four key objectives of introducing countercyclicalbuffers are:

• Dampening any excess cyclicality of the minimum capital requirement;

• Promoting more forward-looking provisions;

• Conserving capital to build buffers at individual banks and the banking sector to be used intimes of stress;

• Achieving the broader macro prudential goal of protecting the banking sector from periods ofexcess credit growth.

In the annex (published in July 2010), BCBS states that the capital conservation buffer should be availableto absorb losses during a period of severe stress while the countercyclical buffer would extend the capitalconservation range during periods of excess credit growth (or other appropriate national indicators).

Through a quantitative impact study, BCBS is considering ways to mitigate cyclicality, by adjusting for thecompression of probability of default estimates in the IRB approach during benign credit conditionsthrough the use of downturn probability of default estimates.

The Proposal provides that a buffer would be “deployed when excess aggregate credit growth is judgedto be associated with a build-up of system-wide risk to ensure the banking system has a buffer of capitalto protect it against future potential losses.” Accordingly, such countercyclical capital buffers are expectedto be deployed in a given jurisdiction only on an infrequent basis,“perhaps as infrequently as once every10 to 20 years.” In general, national bank regulators would inform banks 12 months in advance of theirjudgment of any necessary “buffer add-on” in order to give banks time to build up the additional capitalrequirements, while reductions in a buffer would take effect immediately to help reduce the risk that thesupply of credit would be constrained by regulatory capital requirements.

Under the Proposal, internationally active banks would look at the geographic location of their creditexposures and calculate their buffer add-on for each exposure on the basis of the buffer in effect being inthe jurisdiction in which the exposure is located. (In other words, an internationally active bank’s bufferwould effectively be equal to a weighted average of the buffer add-ons applied in jurisdictions to whichit has exposures.) Accordingly, internationally active banks “will likely find themselves carrying a smallbuffer on a more frequent basis, since credit cycles are not always highly correlated across the jurisdictionsto which they have credit exposures.” The Proposal also notes that the BCBS is continuing to consider thehome-host aspects of the Proposal.

13 The issue of procyclicality was specifically addressed by the BIS in the Working Paper:Countercyclical capital buffers:exploring options,published on July 22,2010.

28


To assist the relevant national banking regulators in each jurisdiction in making buffer decisions, the BCBSdeveloped a methodology to serve as a common starting reference point. The methodology “transformsthe aggregate private sector credit/GDP gap into a suggested buffer add-on,” with a zero guide add-onwhen credit/GDP is near or below its long-term trend and a positive guide add-on when credit/GDPexceeds its long-term trend by an amount which suggests there could be excess credit growth. The BCBSnoted, though, that national authorities are not expected to rely mechanistically on the credit/GDP guide,but rather are expected to apply judgment in the setting of the buffer in their jurisdiction after using thebest information available to gauge the build-up of system-wide risk.

In the latest September press release the Basel Committee agreed that a countercyclical buffer within arange of 0% – 2.5% of common equity or other fully loss absorbing capital will be implemented accordingto national circumstances. The purpose of the countercyclical buffer is to achieve the broader macro-prudential goal of protecting the banking sector from periods of excess aggregate credit growth. For anygiven country,this buffer will only be in effect when there is excess credit growth that results in a system-widebuild up of risk. The countercyclical buffer, when in effect, would be introduced as an extension of theconservative buffer range. Please see last section of this paper for an overview of the agreed detailedimplementation timeline.

There is a general consensus on the need for stronger counter-cyclical capital buffers to be part of theBasel capital framework. This is clearly a powerful and necessary starting point. However, the challengelies in the calibration of the parameters when trying to implement in practice this generally agreed objective.

We have identified several design issues that we believe deserve further attention by the committee14:

1. Which approach to follow in determining the counter-cyclical capital buffer: discretionary,rule based or mixed?

2. What are the most appropriate policy instruments to introduce counter-cyclicality?

3. Is there a need for a counter-cyclical liquidity measure?

4. What is the most appropriate accounting treatment of a counter-cyclical capital reserve?

5. How do you provide disincentives for the (mis)use of financial innovation and tighten counter-cyclical rules for financial institutions that extensively use it?

6. How do you avoid regulatory arbitrage associated with the introduction of a countercyclical regulatory capital measure?

7. Does size and correlation matter during systemic crisis?

8. Is there a need for a holistic balance sheet management approach?

After a detailed analysis of possible alternative methodologies that can be used in determining thecounter-cyclical capital buffer, three approaches seem to emerge so far:

i. The discretionary approach.

ii. The rule-based approach.

iii. A mixture of the two.

14 More details on this topic can be found in the Algorithmics response to the Basel Committee on the countercyclical capital buffer.

29


With a discretionary system, bank regulators would need to judge the appropriate level of required capitalratios in light of analysis of the macroeconomic cycle and of macro-prudential concerns. It would dependcrucially on the quality and independence of the judgments made. Using a formula-driven system, therequired level of capital would vary according to some predetermined metric such as the growth of thebalance sheet.The Turner and other (e.g., Geneva) reports make the case that there is merit in making theregime, at least to a significant extent, formula driven.This could be combined with regulatory discretionto add additional requirements on top of the formula-driven element if macro-prudential analysissuggested that this was appropriate.We believe this is the right approach to follow.

There are several options that have been analyzed in different consultative papers, research papers andinternational reports (see references at the end of this document) regarding instruments that can be used asa counter cyclical buffer.We believe that the right way forward is to consider a combination of instruments.At a minimum we believe that any countercyclical buffer rules should consider an increase on capitalrequirements, as currently outlined in the July consultative paper.The buffer must be able to absorb losseson a “going concern” basis. Consequently, the buffer must comprise the highest quality capital, mostlikely equity and retained earnings. Also, to avoid regulatory capital arbitrage, as we will describe in thesubsequent sections, we recommend that further countercyclical measures be added, in particular:

• A cap on leverage15 and

• A capital multiplier if significant currency or maturity mismatch is found

As solvency and liquidity are complementary, these rules should be implemented jointly, which wouldimply requiring more capital in a counter-cyclical way for institutions with large maturity mismatches.However, as capital will never be enough to deal with serious liquidity problems, there is a clear case forhaving a counter-cyclical liquidity requirement as well. As the U.S.Treasury September 2009 Report argues,excessive funding of longer term assets with short term debt by a bank can contribute as much or moreto its failure as insufficient capital. Furthermore, the report states that liquidity is always and everywherea highly pro-cyclical phenomenon. Indeed, because capital, even though high, may be insufficient to dealwith liquidity problems in a crisis, sufficient liquidity requirements are also very important, and need to bedetermined simultaneously with the general capital requirements in an integrated/out-of-silos framework.Banks with larger structural funding mismatches, or those that rely on volatile short-term funding sourcesshould be required to hold more capital.This would force the banks to internalize higher liquidity risks asa cost, thus encouraging them to seek longer term funding.The Geneva Report and Warwick Report gofurther by recommending that regulators increase the existing capital requirements by two multiples,one linked to the growth of credit, and the other to maturity mismatches.

The accounting treatment of a counter-cyclical capital reserve is hugely important. As regards to accountingdisclosure rules, these should satisfy both the needs of investors and those of financial stability.An optimalapproach may be to rely on dual disclosure, where both current profits and losses are reported, and profitsafter deducting a non-distributable counter cyclical buffer that sets aside profits in good years for likelylosses in the future.

15 For a detailed analysis of this topic please refer to the leverage ratio section 3.3

30


Financial innovations increase during booms, when new and untested instruments that are difficult tovalue are introduced.This exacerbates pro-cyclicality, as new, often opaque and complex instruments canhide and under-price risk. Regulators should introduce appropriate model risk capital charges if suchinstruments are traded, or at least tighten counter-cyclical rules for financial institutions that extensivelyuse them.When new and complex products are originated and then distributed it is non-trivial to understandthe interplay between risk classes, regulatory and accounting treatments and how, ultimately, this cocktailwill influence and drive the setting up of the overall business strategy.The evolution of the collateralizeddebt obligation (CDO) market during the credit crunch is a good example of this. Only through a holisticview of the balance sheet is it possible to disentangle and understand if the resulting “new”bank portfoliois aligned with the overall risk appetite of the bank.

The standards proposed in the consultative document can be seen as an effort by the Basel Committee toachieve a higher degree of harmonization among supervisory regimes for countercyclical capital bufferrequirements. We strongly endorse this effort. We believe that as much effort as possible should be putinto the convergence of local supervisory regimes of capital and liquidity risk supervision. To avoidregulatory arbitrage, the comprehensiveness of counter-cyclical regulation is an important issue, bothnationally and internationally. The best approach utilizes equivalent comprehensive counter-cyclicalregulation for all institutions, instruments, and markets.This would include also all non-banking financialinstitutions, such as hedge funds, private equity, insurance companies etc (the so-called shadow bankingsystem), as well as all instruments within banks – by consolidating all activities onto the balance sheet.It should also include counter-cyclical margin and collateral requirements on all securities and derivativesinstruments. Having different capital buffers in different jurisdictions will contribute to differences in costof capital. This might encourage regulatory arbitrage where multi-national companies will borrow fromthe cheapest jurisdiction that has a lower counter cyclical buffer to finance activity in other jurisdictions.

The emphasis that the U.S. Treasury report and other reports place on higher capital requirements forsystemically important institutions draws on research from the BIS and elsewhere showing that largebanks,and those more exposed to system-wide shocks,contribute more than proportionally to systemic risks.Both the size of individual banks and the total banking system – or even financial system – are important,because in crisis situations they may need to be bailed out. In addition to size, bank business modelsshould also be taken into account. For example, a bank largely financing exposures through depositsshould have a smaller countercyclical buffer relative to a bank that is completely dependent onwholesale funding.

Supervision and business strategy is a multidimensional discipline: increasingly complex balance sheets,opaque structured products embedding unclear leverage and optionalities all make it difficult to gain anoverall, timely, transparent and objective view of the interplay among risk types. It is therefore not alwaysclear what exact business strategy is being pursued at a legal entity or even at the holding group level.

31


The real problem has become that the resulting overall bank portfolio balance sheet at legal entity levelis now very difficult to understand. As a result, communicating the business strategy to the interestedinternal and external stakeholders, and how this is aligned with the overall risk appetite of the financialinstitutions, has been, and still looks to be, a challenge. This means that bankers need to use new “fit forpurpose” tools and methodologies that can identify the real risk drivers, their interplay, and the role theyplay within different legal jurisdictions, accounting, and under differing regulatory treatments. The limi-tations of the current regulatory, organizational and business silos mindset is probably the biggest andtoughest lesson learned from the crisis.We strongly suggest that regulators endorse and gradually intro-duce, in the spirit and in the letter of the upcoming legislation, this holistic view of the balance sheet,where the interplay of liquidity risk and economic capital is more precisely described. Consequently, werecommend that the impact on the bank and the financial system as a whole be estimated once the stresstests are simultaneously and coherently (same time horizon and severity) executed across all risk types. Webelieve that the need to simultaneously look at the whole legal entity at balance sheet level (and at itsdynamic evolution under stress conditions) from a risk, economic, regulatory, and accounting perspectiveis not a sophistication but a necessity.

3.5. Systemic Risk

One of the findings of the Committee is that while the interconnectedness of international banks hassupported economic growth, in time of distress this interconnectedness transmits negative shocks acrossthe financial system and the economy. The Committee is in the process of developing policy optionsdesigned to reduce risks related to the failure of systemically relevant, cross–border institutions. Theseoptions include empowering the regulatory authorities to write-off or convert certain capital instrumentsinto common shares when a bank becomes unviable, the introduction of a capital and/or liquiditysurcharge for cross-border institutions, or the introduction of contingent capital as an element of thecapital base. These proposals are expected to be issued for draft consultation in December 2010.

32


4. Implementation TimelinesAt its 12 September 2010 meeting, the Group of Governors and Heads of Supervision fully endorsed theagreements reached on 26 July 2010. The capital reforms, together with the introduction of a globalliquidity standard, deliver on the core of the global financial reform agenda and will be presented to theSeoul G20 Leaders summit in November. Annex 2 of the document, reproduced below, summarizes thekey requirements and the required implementation deadlines.

5. Business impact and challenges: exploring the interplaybetween Liquidity and Capital

The Basel II Accord,which was ratified in 2004,had as its main objective increasing the risk sensitivity of capitalrequirements while maintaining the capital amount at a systemic level. Basel II focused on calculatingminimum capital requirements and postponed the discussion on capital quality and definitions. It accountedfor liquidity risk through Pillar 2, while discussion of liquidity risk had not matured by the time the financialcrisis hit. It was liquidity risk that led to the failure of financial institutions like Northern Rock and Bear Sterns.

Also, the Basel II accord did not consider leverage in determining the capital requirements of a bank. Somemajor banks, while being well capitalized from a Basel II perspective, had leverage ratios of 70:1 withouttaking into account netting (30:1 when netting is taken into account). There was a lot of faith in the risksensitivity rules under Basel II. However, some of this was misguided.The capital requirements and the riskmeasurement mechanisms, especially in the trading book, led to regulatory arbitrage. For example,securitization transactions got much more lenient treatment in the trading book versus the banking book,and these shortcomings in the Basel II framework in a way contributed to the capital arbitrage betweenthe banking and trading books and provided incentives for banks to assume high-risk trading strategies.Some of the new rules address these issues.For example, the July 2009 Enhancements to Basel Frameworkaddressed the inadequate capital requirements for securitization transactions,and eliminated the possibilityof capital arbitrage from the banking book to the trading book by aligning the capital requirements acrossthe banking and the trading books.

Phase-in arrangements (shading indicates transition periods)(all dates are as of 1 January)

2011 2012 2013 2014 2015 2016 2017 2018 As of1 January

2019Leverage Ratio Supervisory monitoring Parallel run Migration to

1 Jan 2013 – 1 Jan 2017 Pillar 1Disclosure starts 1 Jan 2015

Minimum Common Equity Capital Ratio 3.5% 4.0% 4.5% 4.5% 4.5% 4.5% 4.5%

Capital Conservation Buffer 0.625% 1.25% 1.875% 2.50%

Minimum common equity plus capital 3.5% 4.0% 4.5% 5.125% 5.75% 6.375% 7.0%conservation buffer

Phase-in of deductions from CET1 20% 40% 60% 80% 100% 100%(including amounts exceeding the limitfor DTAs, MSRs and financials)

Minimum Tier 1 Capital 4.5% 5.5% 6.0% 6.0% 6.0% 6.0% 6.0%

Minimum Total Capital 8.0% 8.0% 8.0% 8.0% 8.0% 8.0% 8.0%

Minimum Total Capital plus 8.0% 8.0% 8.0% 8.625% 9.25% 9.875% 10.5%conservation buffer

Capital instruments that no longer qualify Phased out over 10 year horizon beginning 2013

as non-core Tier 1 capital or Tier 2 capital

Liquidity coverage ratio Observation IntroducePeriod Minimumbegins Standard

Net stable funding ratio Observation IntroducePeriod Minimumbegins Standard

Implementation timeline reported in the 12th September press release

33


It is widely believed that the proposed capital and liquidity rules would significantly increase the bankingsector’s capitalization and funding levels. One study by McKinsey titled “Basel III:What the draft proposalsmight mean for European banking”estimates that the industry would need to raise an additional 40% to 50%of its current Tier 1 capital base.Top 16 European banks will need to raise 700 Billion Euros in capital and1.8 trillion Euros in long-term funding. It is also estimated the Industry ROE would reduce by about 5%from its current level of 15 %.

While the efforts by the Basel Committee are believed to result in lower systemic risk and lower risk forthe individual banks over the longer term, the need for the huge infusions of capital in the short tomedium term places a huge strain on the capital markets. As investors perceive that current risk in thebanking system is high, the required rate of return for any investment in the banking sector will be high.This,viewed along with the lower ROE from the banking industry,and the proposed restrictions on earningsdistributions is likely to create a strain in the capital markets till such time the perception of reduced riskin the banking sector sinks in the investors’ minds.

Above estimates assume current bank structures remain the same, but given the scope of the Basel IIIrules large scale restructuring of bank balance sheets and corporate structures are on the cards.For example,banks might reduce deferred tax assets (assuming profitability over the next few years), sell minority-owned subsidiaries and move away from unsecured interbank funding with larger banking groups.Trading books will almost certainly reduce given estimates that capital might increase about three fold;estimates are high as 20-fold for some banks.

Large multinational banks might look more keenly on regulatory arbitrage as capital becomes dearer, bytaking into account the differences in regulatory requirements across geographies while deciding thebooking location for transactions.As long as regulators allow them,banks will also raise deposits in countrieswhere it is cheapest to fund business in other jurisdictions.

The proposed buffers – capital conservation buffer and the countercyclical capital buffer – are likely tosmooth the capital requirement crests and troughs.

Under the Basel II regime international banks were fairly well capitalized and regulatory capital was rarelybinding. However, with Basel III regulatory capital, and for some large banks leverage ratios and liquidityratios, will become binding constraints when making business decisions. Currently most banks look atregulatory and economic capital when making business decisions. With Basel III this process becomesmore complex.Banks will need to examine the incremental impact of deals on regulatory capital,economiccapital, leverage ratios and liquidity ratios, and also how it impacts capital and liquidity buffers.

This will lead to fine tuning of RWA in the bank’s effort for capital conservation. Banks will focus on dataquality with a special focus on credit mitigation. Some banks currently do not centrally collect collateral,guarantee and netting data for smaller obligors. Although these counterparties might be smaller, in totalthis effort will tend to reduce RWA significantly. Also, banks will use credit mitigant optimization routines,where mitigants are eligible across lending facilities to apply the mitigants to minimize total RWA acrossthe lending facility. Similarly, CCF models will also be refined to achieve capital optimization. AlthoughBasel III precludes using rating triggers to reduce EADs in the current period, instituting a rigorous procedurecan reduce CCF estimates in future periods as well as reduce losses in the current period.

34


As regulatory capital becomes binding, banks will move away from capital heavy sectors to capital lightsectors and adjust the business model accordingly. In terms of the product analysis, the capital costs forOTC derivatives, liquidity facilities and short term/long term corporate loans are expected to increase,while the funding costs for OTC Derivatives,fixed income bond investments,covered bonds, liquidity facilitiesand short term retail loans are expected to increase.Therefore,there will be a move away from these products.For example, retail mortgages get better capital treatment than commercial mortgages. It is noted withinterest that although the BCBS has increased capital on the capital markets side of the business, there isno language to require banks to strengthen underwriting procedures for retail mortgages and other retailloans. Regardless, in most countries banks have strengthened underwriting procedures, and strict rulesregarding securitizations will reduce the incentive for banks to be lax on this front. This will cut off thevicious cycle of originate to securitize.

Basel III is highlighting the integral nature of credit, market and liquidity risk. Before the crisis, most banksappeared to be well capitalized, although they were pushed to the brink due to liquidity risk.The followingis a quote from the BCBS,“Findings on the interaction of market and credit risk,Working Paper No.16, 2009:

Studies suggest that banks’exposures to market risk and credit risk vary with liquidity conditionsin the market, and liquidity conditions in turn are also determined by perceptions of marketand credit risk. This suggests that banks and regulators need to think about a framework thatbetter integrates all three types of risk.”

We expect BCBS to smooth out the internal inconsistencies and the double counting in the currentproposals before finalizing the rules. The longer implementation framework gives banks and regulatorstime to optimize the rules as well as not adversely affect the current tentative growth prospects for theeconomy. In particular, a move towards an integrated framework will let banks examine the impact ofdifferent hedging strategies. Some strategies might reduce credit risk but increase liquidity risk, for example.It is imperative that banks gain a holistic view of risk.

5.1. Exploring Interconnections and Trade-Offs between Capital and Liquidity

The new rules could repeat the mistakes of the past by compounding the ‘silo’-based approach to riskmanagement.

It is of course right that governments and regulators take stringent steps to ensure we never again findourselves in a situation where billions of dollars of taxpayer’s money is used to save the banking system.But to the question: Are we on the right track with Basel III? The answer at the moment is that it goes onlypart way to addressing the weaknesses of the established ‘silo’-based approach to risk management.Granted there has been a regulatory and business push to break down risk silos in recent years, withmarket and credit risk coming together over the last three to four years.This has been driven by changessuch as the introduction under Basel of the Incremental Risk Charge for market risk and the need to havea centralized credit valuation adjustment (CVA) desk with profit and loss responsibility.The recent emphasison CVA risk is a first step towards some banks managing complex counterparty relationships and theinteraction of market and credit risk in an effective way. Operational risk meanwhile has moved under themarket risk group, mainly because of the analytical nature of measuring and managing operational risk.However, the systems still tend to be separate.

35


But these are minor advances in addressing a pervasive,deeply ingrained problem.As the banking businesshas become more complex over the last 30 years, banks in general have tended to take a bottom-up,piecemeal approach to developing risk management functions.When it comes to risk measurement andmanagement, the default option has been for banks to have separate units managing different types ofrisk. At the same time, as trading and structured finance has become more spread out and complex, thecorporate structure of banks has also become more complex, such that different parts of the business runtheir own profit and loss books. This has made it harder, in general, for managers at the top to have anall-encompassing view of the risks on the group’s balance sheet.

As risk management has come into its own as a primary discipline within banks, reporting tools and riskmanagement functions such as stress testing have not been integrated across different departments.Systems have been maintained mostly by decentralized IT functions supporting finance, treasury,business lines and risk management. There has been less thought given to reconciling data acrosssystems, which has led to benchmark risk management practices, and ultimately a culture of risk manage-ment that has failed to view risk in a necessarily holistic way. A truly effective risk management systemwould take a top-down approach to risk measurement and reporting, viewing and managing the intercon-nections between risk factors at a high level, such that their potential impact on the balance sheet can beproperly accounted for.

It is now clear that this compartmentalized approach, with add-ons to meet minimum regulatory require-ments, was seriously flawed.With the emergence of Basel III we have an opportunity to reverse the trendtowards splitting risk into different silos. However, as things stand at the moment it seems the new ruleswill again fail to produce a framework that refutes the compartmentalized way of thinking.

This is seen in the way the Basel Committee is addressing liquidity risk, and specifically in the lack of recog-nition of the connections between leverage, capital and liquidity.The current stress test recommendations,for example, call for the stress testing of liquidity and capital separately. Witness this summer’s widelypublicized European bank stress test exercise. Although the exercise showed that most banks will be ableto withstand a significant and protracted period of stress, the tests focused on capital levels only.They completely failed to touch on liquidity risk.

If we are to have a robust and truly risk-based framework then the interdependence of capital and liquidityrisk must be addressed.One can see the attractiveness of trying to get to grips with liquidity risk by viewingit largely on a stand-alone basis. But this is a mistake. If we are to have a real sense of the nature of liquidityrisk, and from there be in a position to put in place systems to effectively manage it, then we must beprepared to engage in the more difficult intellectual challenge of viewing and managing risk holistically.

5.2. Misunderstanding How Liquidity Risk and Capital are Connected

By viewing capital as a primary mitigant of liquidity risk we fail to understand the nature of that risk.Capital mitigates unexpected losses, but not cash flow imbalances – i.e. funding liquidity risk. Liquidityrisk is crystallized when a bank has to undertake a last minute fire sale of assets to meet its obligations.In short, if an institution has a liquidity problem then it needs cash, not capital. Indeed, should a liquiditysituation arise and the bank begins using reserves set aside to guard against liquidity risk in order toabsorb losses and meet obligations, then the value of the company and therefore also the value of thecapital is likely to go down, since the bank will start to be perceived as “riskier”. Liquidity risk and capital aretherefore inextricably linked.

36


The Basel Committee’s primary response to liquidity risk, the liquidity coverage ratio and the net stablefunding ratio, do not recognize this link. As with previous compartmentalized approaches to risk manage-ment, these ratios view liquidity risk more or less as a stand-alone risk silo. As such, the implementationof the current approach does not effectively address the flawed silo-based approach.

The prescriptive nature of these ratios is not helpful, as it does not allow the tailoring of a liquidity riskbuffer to the needs of the specific institution. Under a top-down, holistic risk management model, the seniormanagement of the bank would decide on the size of the liquidity buffer and what survival horizon isappropriate for it, based on a careful assessment of the bank’s overall risk appetite.This is important froma best practice governance perspective, because if an institution is holding more than the needed amountof liquid assets then the part of the liquidity buffer that is not needed has an opportunity cost associatedwith it – that money could be deployed elsewhere to make a higher return for shareholders. And if theinstitution holds less than necessary to maintain stability then the bank risks bankruptcy.

6. Technology DirectionOver the last two decades, banks approached risk management from a silo approach across market, creditand operational risk. Basel I was addressed mainly through finance systems. Banks developed market risksystems after the 1996 Amendment. Liquidity risk became a risk discipline only over the past few years.

As risk systems grew organically over the years with own data requirements, reporting tools and stresstest set ups, they addressed requirements of different departments within the banks. Risk managementitself has come into its own as a primary discipline within the banking industry only over the past 15 yearsor so.Systems were maintained mostly by decentralized IT functions supporting finance, treasury, businesslines, risk management etc.There was no thought given to reconciling data across these systems.

Spurred by Basel II, larger banks embarked on enterprise data warehouse projects to collect data acrosstrading books, corporate, retail and securitization exposures. However, these systems originally focusedon driving credit risk-weighted asset calculations, local regulatory reporting and Basel pillar 3 reporting.Over the last three to four years banks also invested in expanding these data warehouses to cover crediteconomic capital. Market risk and operational risk were calculated separately and typically broughttogether at the reporting layer, with manual interventions for regulatory and management reportingpurposes under Basel II.

With the advent of ICAAP stress test requirements banks needed to examine the impact of a market stressevent across risk silos. Banks realized the shortcomings of current systems, but manual aggregation wasa practical solution given these stresses needed to be run typically only at a semi-annual frequency.However, the systems are prone to manual errors and find it difficult to cope with ad-hoc stress testingrequirements that regulators are now pushing for.

37


To address Basel III, banks will need to re-think data and IT strategy.The main drivers will be:

• The ability to reconcile data across different risk categories.This will lead to a system based oncommon data inputs to drive market, credit and liquidity risk. Position and counterparty/obligordata should be driven from a single source of truth. A single data load with all the attributesrequired for market, CCR, RWA, economic capital and liquidity risk should be extracted fromsource systems. Since this data will be shared across risk types, data reconciliation requirementswill be automatically met.

• Drive the system with common risk factors to enable consistent stress testing across market,credit and liquidity risks. In addition to stress testing, this will also let the bank look at the impacton its capital and liquidity position of changes in assumptions. For example, it would let thebank examine the impact of an increase in the margin period of risk when there are over 5000transactions in the netting set.

• Consistent calculation engines that share common models and provide consistent measuresacross risk types. For example, the pricing models used for market risk valuation purposesshould be used for counterparty credit risk simulations. Cash flow generation for liquidity riskshould also leverage the cash flow generation routines that should be common to market riskand CCR. On the capital front, using consistent data and models will allow the user to breakdown the differences between regulatory and economic capital into sources of risk, such asname concentration, sector concentration, migration risk etc. An economic capital model thatallows the user to configure economic capital calculations under multiple assumptions (e.g.full granularity, default/no default mode etc) will enable such a breakdown of sources of risk.

• Integrated reporting across risk types, which is the ultimate aim of the system.This would allowsenior managers and investors to get a consistent view across the enterprise of the impact ofdifferent types of risk. There will be demand for systems than can perform “what-if” analysisbased on incremental transactions or scenarios. This will enable business users to get acomprehensive view of the risk of incremental trades or hedging strategies. For example, aparticular hedge might reduce credit risk but increase liquidity risk.

• Systems should allow both large volumes, enterprise-wide batch runs while allowing forinteractive what-if analysis.This will let the bank examine the impact across market, credit andliquidity risk of incremental deals. It will also let the bank check the impact of new regulationsor assumptions such as the margin period of risk.

It will make sense for banks to leverage existing investment in Basel II systems to achieve this. Banks thathave been successful in building enterprise data warehouses will want to expand them to address BaselIII.We see a trend where banks extend the current market risk system for CCR mainly to ensure consistentpricing models. As computation has become cheaper it is feasible to use the same market risk pricingmodels for counterparty credit risk simulations.

At most banks RWA and economic capital systems are now driven by the same data. Liquidity risk coversthe enterprise and uses market risk factors for cash flow generation and stress testing.

The following table summarizes the current silos of risk types,their scope,risk factors, functionality,measures,confidence intervals and primary functionality.

Market CCR RWA Credit Risk Liquidity

Scope Trading book Trading book Enterprise Enterprise Enterprise assets assets assets assets assets and

liabilities

Risk Factors IR, FX, IR, FX, Equity, PD, LGD, PD, LGD, IR, FX,Equity, Commodity, EAD, M EAD, Equity,

Commodity, Volatility etc., Transition Commodity,Volatility etc. creditworthiness Matrix, Volatility etc.,

for CVA Correlation cash flowMatrix generation

Measures VaR, IRC PFE at any CI, RWA, EL UL, EL, CBC, FLE,EPE, CVA survival LVaR, LCR;

probability NSFR;survivalhorizon

Time horizon 10-day 1-year for EPE. 1-year 1-year for Short Term:Multi-year for PFE ICAAP; daily up to

3 and 5 year 30-90 daystypically for Long Term:

capital yearly up to planning 15-20 years

Confidence 99% 50% for EPE, 95 99.9% 99.9% for Non level or 99% for PFE regulatory Stochastic

purposes; Stress Testbased on riskappetite for

internal

Main Stress Simulation Undrawn Additive Survivalfunctionality testing through time allocation, measures, Horizon &

incorporating all mitigant Reverse Liquiditycredit mitigation optimization, stress tests Buffer

including netting, retail pooling,collateral etc. reconciling to

GL, stress testing

38


As can be seen, different types of risks should be driven by common data. Risk types might focus on partof the book or the enterprise, and might need incremental data to drive calculation engines. Providingunderlying data from a single source of truth will result in a consistent, reconciled system while providinglong term cost savings in terms of reduced manual intervention.

The thrust in the future will be towards an integrated risk management platform for regulatory purposesthat will allow for ad-hoc stress tests.The original Basel systems for credit risk moved from finance to capitalmanagement or risk management due to Basel II. With Basel III, the trend is towards integrated systemsmoving to enterprise risk groups that will look at risk holistically.

39


7. Conclusion: Moving Towards a Holistic SystemIt is clear that we need to break down the silo-based approach to managing risk and that Basel III is thebest opportunity we have of doing it. That starts with the senior management establishing a detailed,clearly defined definition of the overall risk appetite of the bank. This ensures that shareholders, depositholders and other stakeholders have a clear understanding of the business strategy. From there, a numberof steps should be taken to ensure that the bank truly takes ownership of the risks it is running at thegroup level, as well as at lower business or division levels.

Systems should be developed based on common data inputs to drive market, credit and liquidity risk.A single data load with all the attributes required for market, counterparty credit risk, RWA, economiccapital and liquidity risk should be extracted from source systems. Since this data would be shared acrossrisk types, data reconciliation requirements would be automatically met. At the same time, there should beconsistent calculation engines that share common models and provide coherent measures across risktypes. For example, cash flow generation for liquidity risk should use the same cash flow generationroutines common to market risk and counterparty credit risk.

There should also be integrated reporting across risk types to give senior managers and investors aconsistent view across the enterprise of the impact of different types of risk. Meanwhile, systems shouldbe designed that allow both for large volume, enterprise-wide batch runs and also interactive ‘what-if’analysis.This will also enable consistent stress testing across market, credit and liquidity risks, as they willall be driven by common risk factors. The implementation of such measures would allow the interplaybetween capital and liquidity to be fully tested.

With this breaking down of risk silos,senior management will be able to view a ‘dashboard’of risk indicatorsthat give them a true picture of their group balance sheet and variances from the stated risk appetite.Thiswill mean that senior managers will once again, like the days before the emergence of complex banking,take ownership of the bank portfolio balance sheet at the legal entity level, in turn allowing them to takea harder line if they feel they have to.

As Basel III progresses it is crucial that the interconnected nature of the risks on the balance sheet areproperly assessed, while taking account of regulations and accounting standards. It is, therefore, time tobreak down those silos.

40


REFERENCES1. Algorithmics’Response to the Basel Committee's request for comments on the consultative document:

Proposed Enhancements to the Basel II Framework, April 2009http://www.algorithmics.com/EN/media/pdfs/Algo-GC0409-LtrBaselCom.pdf

2. Algorithmics' Response to the FSA's CP 09/13 – Strengthening Liquidity Standard 2: LiquidityReporting, June 2009http://www.algorithmics.com/EN/publications/whitepapers/registration.cfm?code=wp39

3. Algorithmics’Response to the Basel Committee’s request for comments on the consultative document:International framework for liquidity risk measurement, standards and monitoring, April 2010http://www.algorithmics.com/EN/media/pdfs/Algo-GC0410-LtrBaselCom2.pdf

4. Algorithmics whitepaper:Towards active management of counterparty credit risk with CVA, 2010http://www.algorithmics.com/EN/publications/whitepapers

5. Algorithmics whitepaper: Credit Value Adjustment and the changing environment for pricing andmanaging counterparty credit risk, 2010.http://www.algorithmics.com/EN/publications/whitepapers

6. APRA’s prudential approach to ADI liquidity risk, Discussion Paper, 11 September 2009

7. F. Battaglia and M. Onorato, Liquidity Risk: Comparing Regulations Across Jurisdictions and The Roleof Central Banks, December 2007;http://www.algorithmics.com/EN/publications/whitepapers

8. F. Battaglia, M. Onorato and S. Good, Liquidity Risk Management – Assessing and Planning forAdverse Events, December 2007;http://www.algorithmics.com/EN/publications/whitepapers

9. Basel Committee for Banking Supervision, Principles for sound liquidity risk management andsupervision, September 2008

10. Basel Committee of Banking Supervision,“Findings on the interaction of market and credit risk,Working Paper No. 16, 2009:

11. Basel Committee of Banking Supervision, Strengthening the Resilience of the banking Sector,December 2009

12. Basel Committee of Banking Supervision, International Framework for Liquidity Risk Measurement,Standards and Monitoring, December 2009

13. Basel Committee of Banking Supervision, Strengthening the Resilience of the banking Sector,Annex, July 26, 2010

14. Basel Committee of Banking Supervision Working Paper: Countercyclical capital buffers: exploringoptions, published on July 22, 2010.

15. Basel Committee of Banking Supervision, Group of Governors and Heads of Supervision announceshigher global minimum capital standard, Press Release, Sept 12, 2010

16. Mario Draghi, Chairman of the Financial Stability Board, Letter to the G20 Meeting in Toronto, July 2010

17. European Banking Supervisors, Guidelines on Liquidity Buffers & Survival Periods, 9 December 2009

18. Financial Stability Forum Report, Addressing Procyclicality in the Financial System. April 2009.

41


19. The Financial Times, Move to reassure banks on tough rules, 5 July 2010

20. Geneva Report on the world economy,The Fundamental Principles of Financial Regulation, May 2009

21. Institute for International Finance, Interim Report on the Cumulative Impact on the Global Economyof Proposed Changes in the Banking Regulatory Framework, June 2010

22. McKinsey on Corporate & Investment Banking,Basel III: What the draft proposals might mean forEuropean banking, Summer 2010

23. M. Onorato, A Face-off On Funding: ERM and Liquidity Risk, 2008,TH!NK Magazine; www.algorith-mics.com

24. M. Onorato, Liquidity Before and After:The Emergence Of Balance Sheet Risk Management; June2009,TH!NK Magazine; -Web: www.algorithmics.com

25. M. Onorato, Grasping at shadows: Identifying an effective framework for liquidity risk management,August 2009, GARP Professional;http://www.garp.org/news-and-publications/2009/august.aspx

26. M. Onorato, A Comprehensive Framework For Defining Risk Appetite , forthcoming 2010; TH!NKMagazine www.algorithmics.com

27. Goldman Sachs , Strengthening the Resilience of the banking Sector, Response to the BaselCommittee of Banking Supervision from Goldman Sachs, April 16, 2010

28. The Turner Review: A regulatory response to the global banking crisis, March2009, FinancialServices Authority.

29. U.K.Treasury Committee. (2009). Banking Crisis: regulation and supervision. July, House ofCommons,Treasury Committee, Fourteenth Report of Session 2008–09, London.

30. U.S.Treasury Department. Financial Regulatory Reform. A New Foundation: Rebuilding FinancialSupervision and Regulation. June 2009, U.S.Treasury Department, Washington, DC.

31. U.S.Treasury Department. Principles for Reforming the U.S. and International Regulatory CapitalFramework for Banking Firms. September 2009, U.S.Treasury Department, Washington, DC.

32. Warwick Commission on International Financial Reform: In Praise of Unlevel Playing Fields;November 2009

33. Oliver Wyman, State of the Financial Services Industry Report, 2009

About AlgorithmicsAlgorithmics is the world’s leading provider of enterprise risk solutions.Financial organizations from around the world use Algorithmics’ software,analytics, and advisory services to help them make risk-aware business decisions, maximize shareholder value, and meet regulatory requirements.Supported by a global team of risk experts based in all major financial centers,Algorithmics offers proven, award-winning solutions for market, credit, andoperational risk, as well as collateral and capital management. Algorithmics isa member of the Fitch Group.

www.algorithmics.com

© 2010 Algorithmics Software LLC. All rights reserved.You may not reproduce or transmit any part of thisdocument in any form or by any means, electronic or mechanical, including photocopying and recording,for any purpose without the express written permission of Algorithmics Software LLC or any other memberof the Algorithmics’ group of companies.

ALGORITHMICS, Ai Logo, ALGORITHMICS & Ai Logo, ALGO, MARK TO FUTURE, RISKWATCH, KNOW YOURRISK, ALGO RISK, ALGO MARKET, ALGO CREDIT, ALGO COLLATERAL, ALGO FIRST, ALGO ONE, ALGOFOUNDATION, ALGO FINANCIAL MODELER, ALGO OPVAR and TH!NK Logo are trademarks of AlgorithmicsTrademarks LLC.

Documents

Whats New in Basel III