||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
VO L U M E 2 , I S S U E 4 O C TO B E R 2 0 1 2

Intelligent risk KNOWLEDGE FOR THE PRMIA COMMUNITY

EDITORIAL BOARD IN THIS ISSUE

EDITOR EDITOR’S NOTE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Douglas Ashburn
LETTER FROM THE DIRECTOR OF EDUCATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
[email protected]
NEW BOARD MEMBERS ANNOUNCEMENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
PRODUCTION EDITOR
Andy Condurache VISIONS OF RISK
[email protected] The Forward/Future LCR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
The Troubled Industry of Risk Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
REGULAR FEATURES
Risk As An Agent of Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
VISIONS OF RISK Modelling Credit Spreads for Counterparty Risk: Mean-Reversion is Not Needed . . . . . 13
PRODUCER AND EDITOR
Bob Mark SPONSOR ARTICLE
[email protected] Reverse Stress Testing From A Macroeconomic Viewpoint:
Quantitative Challenges & Solutions for its Practical Implementation . . . . . . . . . . . . . . . . . . 16
ACADEMIC PARTNER PROFILE
Andy Condurache CHAPTER REPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
[email protected] PRMIA Philippines
CHAPTER REPORT WHAT’S ON THE WEB? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Kristin Lucas Risk Management and Governance of Financial Institutions
[email protected]
LEARNING OPPORTUNITIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

ACADEMIC PARTNER PROFILE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

Kent Business School, University of Kent

PRMIA LEADERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Board of Directors

Thanks to our sponsor, the exclu- SPONSORED BY:

sive content of Intelligent Risk is
freely distributed worldwide to Moody’s Analytics helps capital markets and risk management profes-
over 87,000 PRMIA members. If sionals worldwide respond to an evolving marketplace with confidence.
you would like more information The company offers unique tools and best practices for measuring and
about sponsorship opportunities managing risk through expertise and experience in credit analysis, eco-
contact [email protected]
nomic research and financial risk management. By providing leading-edge
software, advisory services, and research, including the proprietary analy-
sis of Moody’s Investors Service, Moody’s Analytics integrates and cus-
www.prmia.org/irisk tomizes its offerings to address specific business challenges. For more
information please visit: www.moodysanalytics.com/RSTmacro

©2012 Professional Risk Managers’ International Association, All Rights Reserved

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
EDITOR’S NOTE

THE CHANGES COME IN THREES

DOUGLAS ASHBURN, EDITOR

W
e have all heard the old adage that significant changes come in sets of
three. Prudent risk professionals, of course, know that no such link exists
among independent events, and thus tend to pay no heed to them. Yet,
for this issue of Intelligent Risk (iRisk), I must point out that there have been exactly
three changes made to the editorial board.
First, Carol Alexander has completed her tenure as executive editor. Second, PRMIA would like to welcome Andy
Condurache as the new director of exams and publications. Third, I have agreed to step in as the new editor of
Intelligent Risk, replacing Michael Martin — whom, I hear, is irreplaceable — but I will certainly try my best to fill his
shoes. Best wishes to outgoing members of the editorial team, and good luck to the new team.
One aspect of iRisk that has remained unchanged is the high quality of submissions from PRMIA members and
the risk community-at-large. The current issue features another diverse selection of risk topics from academics,
investment bankers, researchers, and even a central banker. Here is a sample:
■ Banks, bank regulators, and even politicians have spent considerable time and effort (and more than a bit
of lip service) on the stress testing of banks to alleviate systemic uncertainty. But, what about reverse stress
testing, i.e., beginning with the assumption of business failure, and working backwards through the
conditions that would lead to such failure? Two researchers from Moody’s Analytics, Juan M. Licari, and
José Suárez-Lledó, explain the concept and offer a real-world example from the U.K.
■ Implementation of Basel III provisions, while imminent, has thus far left many unanswered questions in the
minds of risk managers. For example, the calculation of the liquidity coverage ratio (LCR) is fairly
straightforward. Yet forward LCR projections, which are not mandatory but would seem to be important
exercises to undertake nonetheless, will require extra thought. Liquidity Risk’s Robert Fiedler takes a look
at how such a process may be structured.
■ Jonathan Howitt takes us on a journey through the history of inherent risk assessment and asks whether
we are truly better off than we were 20 years ago. You may be surprised at the answer.
■ William Mason, who heads the risk department at the Central Bank of Ireland, walks us through the design,
planning and implementation of PRISM, its new risk-based supervision software suite.
■ And, finally, in this month’s PRMIA chapter report we see that, whether one is a tourist or a risk manager,
there is FUN in the Philippines.

In closing, I would like to thank the PRMIA production team for allowing me the opportunity to be a part of this
fine publication. I would also like to thank outgoing editor Michael Martin, not only for his dedicated service to
PRMIA, having edited Intelligent Risk since its inception, but also for imparting me with a few kernels of wisdom
regarding the editorial process. Happy reading!

DOUGLAS ASHBURN, Editor

Douglas Ashburn is a Sustaining Member of PRMIA and a self-proclaimed “trader-masquerading-as-a-journalist.” Doug is a 20-
year veteran derivatives trader and manager of portfolio risk, whose focus has been in foreign exchange options, futures and OTC
derivatives. He is currently Editor-at-Large of John Lothian News, a Chicago-based financial media company, where he regularly
writes a column on FX, edits a managed futures newsletter, and leads the development of MarketsReformWiki, a site dedicated to
tracking and archiving of all information related to global financial regulatory reform.

2 INTELLIGENT RISK E D I TO R ’ S N OT E
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
EDUCATION CONTINUUM

EDUCATION: YOUR OWN PERPETUAL BOND

ALEXANDRU VOICU, DIRECTOR OF EDUCATION

S
ome argue that hindsight is an exact science while others say that it plays tricks
on our minds. In truth, like so many of the regulations being debated today,
hindsight is a double-edged sword. It is, however, difficult to argue with the fact
that, pre-2007, the financial sector suffered from hubris. The industry had embraced the
science that was proving it right, but had forgotten the assumptions behind the models.
For this reason, we were always of the opinion that a good risk manager needs to be in strong command of
mathematical skills, and have such skills instilled through the PRMTM designation. This is true not because
quantification is the solution to everything, but rather because the understanding of models and their limitations can
only lead one to the path of informed decision-making.
Ever since Pascal and Fermat bred the theory of probabilities, we have attempted to measure risk. Finance,
however, is a bit different from the physical sciences, where the fundamental constants are exactly that — constants.
The law of gravity has not changed, nor is such a change expected any time soon, whereas the risk premium in finance
has always been the subject of controversy. In addition, the gravity of finance — volatility — actually has its own
volatility. Other factors such as geographical and social mobility are different across the sphere and in perpetual
motion. The worldly atoms, humans, do not move in a Brownian fashion, but we think, often in quite irrational ways.
If these were the physical laws of the universe we inhabit, we would probably experience a “big bang” every ten years,
if not every day.
Richard Feynman would be much more critical of the social sciences, but I think we are worthy of the chance. If
we can only find the missing common sense, and uphold the trait of epistemological humility in the face of
complexity, we will have proven that social sciences are of great value to society. Humility opens us up to learning
about changing paradigms, and we have all witnessed the shattering power of the status quo.
Fortune favors the prepared. As an analogy, Michael Phelps set a world record at the Beijing Olympics in 2008
with a goggle malfunction that effectively blinded him for the second half of the race. He could not have won without
possessing his own “recovery and resolution plan.” In order to prepare for such a potential mishap, Phelps always
conducted a portion of his training in the dark to simulate a “blind swim.”
Similarly, a champion risk manager must think about risk in an integrated fashion. She must be able to tackle the
multi-faceted nature of risk management with tools, knowledge, experience and a holistic understanding. This is
especially true in a crisis scenario when time, a precious commodity otherwise, is worth exponentially more.
PRMIA is all about being open-minded relative to risk management. We try to keep our members continually
informed about the latest developments in the field, with chapter events, webinars and publications. We have created
the world’s first budget sensitive continuous professional development program with the corporate membership. This
reflects the new reality in which management, and implicitly risk management, will be a combined top-down/
bottom-up process where we create virtuous informational feedback loops to aid in the decision-making process.
Please take a moment to review our program and consider steering your institution in this direction.
Webinars remain the core educational resource within the corporate membership. In past webinars, we have had
the honor of hosting risk management luminaries including Nobel laureates, practitioners, researchers, regulators
and consultants.
We believe that an increased sense of community will lead to the advancement of the profession, just as a country
that opens its economy and migrates towards inclusive institutions lifts its people out of poverty.
Please join us in our endeavor to reduce systemic risk and increase the social role of banking. That is a bond that
will pay in perpetuity. I would like to thank all past, present and future PRMIA volunteers for their wonderful ideas and
sustained effort in making such progress possible.

ALEXANDRU VOICU, Director of Education

L E T T E R F R O M T H E D I R E C TO R O F E D U C AT I O N O C TO B E R 2 0 1 2 3
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ANNOUNCEMENTS

PRMIA MEMBERS ELECT NEW BOARD OF DIRECTOR MEMBERS

C ongratulations to the newly elected Board members, all of whom were elected to 3-year terms. Thank you to the
PRMIA members who voted in this year’s election.

New Board members include:

We also welcome Matthew Chen, Vice

President and Head of Banking at SunGard in
Moorad Choudhry China. The Board named Matthew as the
Treasurer, Corporate Banking Division, successor to Shaun Bond, who was recently
Royal Bank of Scotland informed of a career and location change that
will take him out of Asia, thereby making him
unable to serve the remainder of his three-
year term as a representative for the Asia/Pacific Region on the
Board. Thank you to Shaun for his past year of service, and best
Colin Lawrence (re-elected to his second term)
Director, Risk Specialists Division, Financial
of luck to him in his new role.
Services Authority (FSA)
Our sincere thanks and appreciation to outgoing Board members
Bud Haslett and Carol Alexander for their years of service and
leadership to PRMIA. We wish them every success for the future.

Faruk Patel Thank you also to all of this year’s nominees. We look forward
Manager, Investment Risk, Alberta Investment to their continued participation and leadership.
Management Corporation (AIMCo)

PRMIA Corporate Membership gives your team:

■ Weekly thought leadership webinars led by academics,
practitioners, regulators and authors worldwide
■ Free access to most PRMIA chapter events
■ Free access to risk publications and online resources
■ PRM and Associate PRM discounts on exams and support materials
■ Acknowledgement of your firm’s support on PRMIA website
Empower your staff by
■ HR access to our website for recruitment
joining PRMIA as a
Contact Alex Voicu at [email protected] for more information.
Corporate Member

BECOME A PRMIA SUSTAINING MEMBER

Benefits of Sustaining Membership: Sustaining members receive all of
■ Free access to thought leadership webinars these valuable benefits for a small
annual fee. For further details,
■ Discounts on select PRMIA publications, exam vouchers and online courses including concessionary rates for
■ Discounts on PRMIA events and training courses (up to $100 per course or event) students and low-income groups,
visit bit.ly/PRMIAMembership or
■ Full access to PRMIA Exclusive Content, including surveys, meeting replays and contact [email protected].
PRMIA’s  Jobs Board

4 INTELLIGENT RISK ANNOUNCEMENTS

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
VISIONS OF RISK THOUGHT PIECES FROM PRMIA LEADERS

THE FORWARD/FUTURE LCR

ROBERT FIEDLER

O ne of the provisions of Basel III, the Liquidity Coverage Ratio (LCR), sets a bank’s potential cash outflows in rela-
tion to its capacity to counterbalance them by creating hypothetical inflows from assets which are believed to be
repoable or saleable. The implementation of the LCR in a bank seems to be a straightforward exercise, which can some-
how be seen as decoupled from the more sophisticated internal models a bank might use to economically manage its
funding liquidity. If, however, a bank wishes to manage the LCR not only monthly and in retrospective — as mandatory
in Basel III — but on an on-going, forward-looking basis, it will need to simulate its future balance sheet. This is already
very near to economic risk management techniques.

The LCR — Liquidity Coverage Ratio Today the bank’s assets will, by definition, match its liabilities
In the LCR the bank’s total net cash outflows of the first 30 calendar [1]. Going forward in time, assets and liabilities will not mature
days are cumulated in time and compared with the stock of “high- uniformly. If tomorrow’s maturities include more liabilities than
quality liquid assets” (“HLA”). The term “HLA” is used in Basel III with assets, the bank has a negative liquidity mismatch because there
two distinct meanings. First, it describes all unencumbered and non- are more outflows from liabilities than inflows from assets; if fewer
rehypothecated assets of the bank presumed to be easily and quickly liabilities mature than assets, the bank has accordingly a positive
converted into cash without generating substantial losses. Second, it liquidity mismatch. Negative mismatches, however, can only exist
means the amount of liquidity that can be created by liquefying the as forecasts but not in reality as the central bank will cease
HLA within a fixed time horizon of 30 days. The HLA fall into two payments on the bank’s nostro before it ‘turns negative’. Therefore
categories: level 1 assets and level 2 assets; the latter can comprise the bank must ensure that it will be able to create at least enough
up to 40% of the stock. liquidity to counterbalance the forecasted cash deficit. In the LCR,
a qualified asset (HLA) results in a cash inflow by simply
The Total Net Cash Outflows multiplying its available amount with its market price (which may
Total Net Cash Outflows comprise all expected cash flows derived be diminished by a required haircut).
from the outstanding balances of the bank’s assets; respectively
liabilities or off balance sheet (OBS) commitments which mature Lacking Term Structure of the LCR
within 30 days, multiplied with expected run off/draw down rates. In an economic liquidity risk view (the Forward Liquidity Exposure,
The construction of the net cash outflows reflects the fact that or “FLE”), on each day of the time horizon, all forecasted cash
inflows are not under the control of the bank and are thus more inflows and outflows are netted and carried forward to the next
insecure than cash outflows: day’s FLE. In practice, the FLE can fluctuate from day to day and
net cash outflows = outflows – min{inflows; 75% of outflows} thus the minimum balance will not necessarily occur on the last
day. Contrarily, in the denominator of the LCR the maturing cash
The Ratio as an Inequality flows are summed in a way such that:
The LCR is formulated as: (outflows – min{inflows; 75% of outflows})
stock of highly liquid assets which enforces that the resulting net flows are outflows. Consequently,
>100%
total net cash outflows over the next 30 calendar days the forecasted nostro balance descends monotonously until on the
last day of the time horizon the worst balance appears.
Because inflows and outflows are positive numbers the inequality The LCR’s numerator consists of those hypothetical inflows that
can also be written as: stem from an assumed liquification of the HLA assets within 30
stock of highly liquid assets > total net cash outflows days. It is unclear how quickly the assets are to be made liquid, as
the model specifies only what happens on the last day. Since we do
Complying with the LCR requires a balance sheet structure from
not know what happens inside the time horizon, it is left unclear
banks in which potential cash needs within the first month can be
whether the inequality actually holds during the time horizon, and
covered by the liquidity generated by liquid assets. The LCR can
thus whether the liquidity generating ability of the bank has
also be interpreted as a survival period of at least 30 days.
exceeded its forecasted cash needs on each day.[2]
LCR — First Considerations
Static Run-off View — But Hypothetical Transactions
The LCR is specified as a mixture of a balance sheet view and cash
Only existing assets or liabilities are to be considered when
flow view. The complete set of a bank’s transactions — assets,
forecasting the cash flows for the LCR; not yet existing liabilities or
liabilities and OBS commitments — is transformed into a sum of
assets are not regarded. To determine the counterbalancing cash
cash outflows and inflows at the end of the considered time horizon.
inflows, no explicit assumption is made regarding the repoability

T H E F O R WA R D/ F U T U R E LC R O C TO B E R 2 0 1 2 5
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

and/or salability of the HLA. How can an asset be turned into cash to TNCO(t0) := N0+N1+ ... +N30.
without selling or repoing it? Assets (liabilities) which have been
If we extend the calculation of the Ax, Lx, and Nx until tx = t0+30d
generated e.g. yesterday and will only be paid (received) tomorrow
we get the Forward TNCO(tY) = NY+NY+1+ ... +NY+30.
will create ‘future’ outflows (inflows) from today’s perspective;
there is, however, no such concept of ‘forward’ transactions in
Normal and Forward LCR
the LCR.
The result is:

The Forward LCR • LCR(t0) = HLA(t0)/TNCO(t0) — the normal LCR, calculated in

Assume that a manager appointed to steer a bank’s LCR has today t0 (for the period until t30) and
successfully kept the LCR above 100%, and has reported this to the
• LCR(tY) = HLA(tY)/TNCO(tY) — the Forward LCR, calculated in
regulator. Basel III requires reporting on a monthly basis only, but in
tY (for the period until tY+30).
order to steer today the LCR that is to be reported in 30 days, the
LCR-manager needs to be able to forecast now its value then. In a
Degrees of Freedom to Simulate the Forward LCR
first approach, the Forward LCR is calculated with the assets and
If we simulate in tx the Forward LCR, we need to estimate the
liabilities, as they exist now — without any assumptions about new
forward market value m(tY;Am) and the forward haircut h(tY;Am) as
future assets or liabilities.
they will prevail in tY. The assumed liquification value will then be
First, we must formalize the calculation of the ‘normal’ LCR as it
equated as an = Anx[m(tY;Am)-h(tY;Am)]. The forward market
is performed today (t0) for 30 days in the future. Next we extend
value can be derived by using the well-known methods from
its calculation to flexible periods (ty, to tz) in the future.
market and credit risk management. The forward haircut depends
also on the future credit quality of the asset, but can also change
Normal and Forward HLA
with the sentiment of markets. If potential counterparties that
We calculate in t0 the ‘normal’ HLA(t0):
would buy or in-repo the asset have little liquidity themselves (or
• the HLA is constituted by the assets A1, A2, ...,AN are at least insecure about their upcoming liquidity situation), the
from the bank’s balance sheet. haircut could significantly rise independently from a deterioration
of the asset’s credit quality[4].
• today (t0) a market value m(t0;An) and a haircut h(t0;An)
The simulation of redemption values αn for assets depends also
exist for each individual asset An,
on forward credit and market risk parameters. The redemption
An individual asset’s assumed liquification value equates as values λk for liabilities can on the one hand be regarded as fixed
an = Anx[m(t0;An)-h(t0;An)]; and all assumed liquification values (for deposits); on the other hand they might be also parameterized
an sum up to the HLA(t0):= a1+a2+ ... +aN. as the bank could buy back its own debt. What we have thus far
assumed to be unchanged is the composition of the balance sheet
If we calculate the HLA instead at a future day tY, the constituting
with existing assets and liabilities.
assets do not change, but some of them will have already matured.
By introducing liquification values a*j which are set to zero, if the
The Future LCR
asset has already matured before tY, the Forward HLA(tY) equates
The Downsides of the Forward HLA
to HLA(tY) := a*1+a*2+ ... +a*n.
The calculation of the Forward LCR equates the ‘normal’ LCR as it
would be calculated on a future day tY but with the bank’s balance
Normal and Forward TNCO
sheet, as it exists today. In other words, the bank’s assets and
If the bank’s balance sheet is constituted of assets AN+1, AN+2, ...,
liabilities mature as contractually scheduled; neither a new asset or
AN+M (which are not of HLA type) and liabilities L1, L2, ..., LK, we
liability is acquired nor an existing asset is sold until tY.
consider for the calculation of the ‘normal’ TNCO(t0) as of today
The downside of the Forward LCR, although technically correct,
(t0) only those asset-respective liabilities which mature until
is that in practice, the bank will replace maturing assets and acquire
t0+30d.
new assets and liabilities[5] — which makes the Forward LCR
Let αN+1, αN+2, ..., αN+M be the redemption values of the assets and unusable as a prediction of the ‘real’ LCR that will prevail in tY.
λ1, λ2, ..., λK of the liabilities; we define for a future day tx (≤
t0+30d): Simulating A More Realistic Forecast for the ‘Real’ LCR
In order to better forecast the ‘real’ LCR in tY, we need to simulate
• Ax as the sum of all redemption values α of the above assets
the bank’s future balance sheet evolution between t0 and tY as
that mature at tx
realistic as possible.
• Lx as the sum of all redemption values λ of the above liabilities
We can, for example, conjecture:
that mature at tx.
• new HLA assets A#1, A#2, ...,A#K will replace some of the
Applying the 75% rule of the LCR gives the (negative) net cash
maturing HLA assets
flow in tx as: Nx := -Lx + min{Ax;75%xLx}.[3]
• the according refinancing transactions need to be regarded as
Consequently, with tx = 0, 1, 2, ..., 30, the TNCO as of t0 equates
well: L#1, L#2, ...,L#K
6 INTELLIGENT RISK VISIONS OF RISK
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
• new assets A✝1, A✝2, ...,A✝P will replace some of the maturing • Liability driven banks, a small minority, have bigger headroom
other assets to model their balance sheets, although the inherent risk is that
they can only place their funds in assets with inappropriate
• new liabilities L✝1, L✝2, ...,L✝Q will replace some of the maturing
return and/or risk characteristics.
liabilities
• other additional assets or liabilities might be acquired by Conclusion
the bank. The Forward LCR is a technically consistent extrapolation of the
Normal LCR into the future. In practice, however, it is not a suitable
The Future HLA* and TNCO* are calculated with the same
instrument, as it ignores the inevitable change of a bank’s balance
methods as the Forward HLAY and TNC0Y , but with the
sheet in the future. The Future LCR gives a much better forecast how
appropriate set of new future assets and liabilities in tY.
the bank’s LCR will equate — if the renewal assumptions for the
The Future LCR* = HLA*/TNCO* is a more realistic forecast of
balance sheet have been set up appropriately.
the LCR as it will prevail in a future day tY, assuming a development
If the bank decides to set up a Forward LCR, it should consider
of the balance sheet as specified above.
if it would need to restrict the outcomes of the underlying balance
sheet simulations or if it could already apply them for an economic
Constraints of the Simulation
simulation of funding liquidity risk.
The question arises if the above acquisition of assets and liabilities
can be freely simulated or if some restrictions must be regarded:
• The first constraint of the proposed simulation is that the sum
of all ‘living’ assets and liabilities need to match at any point of ABOUT THE AUTHOR
the simulation’s time horizon.
Robert Fiedler, is a leading practitioner of
• Note that in the simple buy-and-hold simulation as it is used
Liquidity Risk Management. He has spent
for the Forward LCR above, it is very unlikely that the amounts
over a decade in the treasury/dealing
of assets and liabilities maturing in tY will be identical: therefore
rooms of several international major banks
the bank’s balance sheet will not balance in tY — which is
like Banque Nationale de Paris and
impossible in reality.
NatWest Markets as a money market liq-
• From the above follows that, if the sum of assets does not equal uidity manager, trading interest rate prod-
the sum of liabilities on one day, the bank needs to acquire the ucts and derivatives where he has headed various ALM divi-
appropriate liabilities or assets to equipoise the balance sheet. sions and served numerous ALCOs.
If there are more liabilities than assets, we can assume that the Robert was Head of Methodology and Policy, Liquidity &
bank can give a loan, or at least deposit the cash surplus at its Treasury Risk at Deutsche Bank where he devised the method-
central bank nostro account — but can this be reversed? ology and successfully managed a project (LiMA), which still
measures and limits the funding liquidity of Deutsche Bank
• Classical asset driven banks will acquire the asset first and then Group. Subsequently, Robert coordinated the ALM and
try find appropriate refinancing — which can go wrong, as we Liquidity Risk Solutions at Algorithmics Inc., Toronto. He was
learned in 2007 and 2008. member of the board and Head of ALM & Risk Development at
FERNBACH Software AG in Luxembourg.
Robert is the Founder of Liquidity Risk Corp. (LRC), where
he advises private banks, central banks and regulators on liq-
uidity risk methodologies and helps to build software solutions
[1] This does however not imply that the bank’s nostro account with the to implement the resulting policies.
central bank is necessarily equal to zero. He is a regular speaker at the Bank of International
[2] In the Basel III Liquidity Framework we find “... banks and supervisors Settlements’ Financial Stability Institute in Basel. Before and
are also expected to be aware of any potential mismatches within the 30- during the peak of the recent crisis he advised consultancies
day period and ensure that sufficient liquid assets are available to meet and various institutions including the ECB on liquidity risk
any cash flow gaps throughout the period” (Basel Committee on Banking measurement methodology.
Supervision, Dec-2010, S. 3-4), without explicitly explaining how this
In the last years he helped banks (e.g. BNP Paribas Fortis) to
should be implemented by the banks.
build group-wide liquidity management solutions including
[3] Strictly speaking, we have to separate the cash flows from the off modules for transfer pricing and Basel III.
balance sheet transactions into inflows and outflows and add them to Ax
Robert holds a Ph.D. in Pure Mathematics and worked for
respectively to Lx before equating Nx.
several years in mathematical research. He was the Frankfurt
[4] For this considerations compare: The Liquefiable Index (LiX) of an director of Professional Risk Managers’ Association and co-
Asset — 2012-04-06, by Darren Brooke, Matthias Küstner & Robert founder of Risk Management Association in Germany.
Fiedler; http://www.liqrisk.com/lix/.
Robert recently published the standard book: Liquidity Risk
[5] Only a dying bank will not generate any new assets or liabilities. Modeling at RiskBooks.
Nevertheless, even a bank in an orderly liquification procedure will need to
‘square’ its cash nostro with the central bank.

T H E F O R WA R D/ F U T U R E LC R O C TO B E R 2 0 1 2 7
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
VISIONS OF RISK THOUGHT PIECES FROM PRMIA LEADERS

THE TROUBLED INDUSTRY OF RISK ASSESSMENT

JONATHAN HOWITT

Inherently Flawed From the Bottom Up

I do not know where or how the methodology of inherent and residual risk assessment first arose. I assume it gained
credence in the audit world when internal auditors were asked to justify their recommendations by quantifying the value
of their control recommendations. Since, obviously, the larger the inherent risk the greater the value of the control, and
since no auditor would ever certify a process as riskless, there would always be some level of residual risk. The concept
of inherent risk suited support management as well: ‘the risk of my IT system failing could cost the firm billions’ fits well
with budgets. It also suited the vendor market: many users using toolsets with simple math displayed on tidy heat map
reports showing the value of the control environment. Who would not like it?

Not me, I am afraid. It is not only spurious math, but it is also the to each business area that contributed.
wrong approach. Let us use a simple analogy to demonstrate why. That certainly does not mean we should not be doing bottom up
Pretend you have decided to learn to drive a racing car. You should risk assessment. Quite the contrary — no sensible manager would
already be aware of the inherent risk. If the racing car had no ever want to be blind to the risks in his processes. However, we
controls, such as brakes or steering, for instance, you would not need to be pragmatic about the qualitative and, frankly, often
drive it, would you? Thus, it becomes a fairly worthless discussion subjective nature of such assessment, especially if conducted on
to quantify the inherent risk — we all know the answer is death. In an inherent and residual basis, where there is typically little or no
the same way, no business can operate without controls, so we sophistication in the quantitative methodologies deployed. We
must assume the controls are in place when assessing risk. But we should use bottom up risk assessment for management purposes
can still consider single or multiple failures, for example, what and extract the risk issues and remediation lessons, but we should
would happen in your racing car if the steering fails at 120mph on not get too bogged down on control benefit calculations or capital
a sharp turn? What if the brakes fail at the same time? How often allocations.
might that happen in 10,000 laps? We can ask the same questions
in the business context to understand tail risk: for instance, what if Self-Assessment and the Fifth Amendment
your payment reconciliation process fails during a period of high I think it was in 1993 that I was required, as a business controller,
volumes. In addition, what if your IT system fails at the same time? to complete a new FDIC internal assessment on behalf of the
Based on experience, how likely is that over any ten-year period? business. It made sense from the regulator’s perspective: since
Let us now assume we have actually had a useful risk discussion they did not always have the resources available to audit, they
with business management and we have avoided all the flawed could require firms to complete their own self-assessments.
pseudoscience and pitfalls around trying to quantify inherent and Granted, it established a moral hazard for the firm, but it provided
residual risk, and we have computed the value of the control the regulator with a starting point if a direct review was required
environment across hundreds of individual processes. Let us later. Of course, many of us in large firms have regularly had to
assume we have collected a series of likelihood and impact figures produce self-assessments for personal performance appraisals; it
at consistent confidence levels for each significant activity. Now, is a tried and tested management technique. The only issue is that
how do we aggregate all of these numbers? people will not knowingly incriminate themselves.
Unfortunately, our bottom up assessments simply will not add That is not to say there is no value in a frank discussion on risk
up, nor do they lend themselves to easy modeling. Here is why: strengths and weaknesses with a business, but no firm wants its
even if they had been implemented with rigorous consistency and dirty laundry hung out in public. In fact quite the reverse — it
objectivity, the complexity of the many to many relationships and becomes a ‘make me look good’ exercise if it is going north.
dependencies between processes, risks and controls is too great. Declaring weaknesses will only be worthwhile if it evidences a need
Thousands of detailed risks and controls will diversify massively for more resources, or if it posits a ‘get out of jail free’ card, should
across the process universe, and the outputs will be highly a problem subsequently occur. Without independence, the risk
sensitive to our assumptions on loss distributions, dependencies assessment process can too easily be gamed.
and correlations. However, let us imagine we have a sophisticated What elements are we assessing? The original thinking was to
modeling tool and we can somehow cope with all of this. The focus on purely causal components of risk — people, systems,
outputs will still be very difficult to work through and explain to organization and external factors, often because there was too little
senior management who not only may not be statistically minded, objective data to assess these elements any other way. Most of the
but also incapable of feeding back the results in a meaningful way available data was process-related and could not take into account

8 INTELLIGENT RISK VISIONS OF RISK

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

“
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

qualitative factors, management Oxley brought with them a different

opinion, or attribute a value on focus and an industry of compliance

”
planned remediation. There is nothing more depressing for operational risk from which it is
Self-assessments are difficult to still struggling to shake free. Despite
keep fresh. When first introduced, for the risk department than being my comments in New York, It is not
senior business management might just a US problem — many major
be directly involved, but the
the reconciliation agent between European and Asian banks are either
following year it gets delegated until business self-assessments and internal US listed or encouraged by their
eventually a contractor, trainee or auditors or self-serving consulting
even intern is simply managing an audit’s risk assessment. firms to at least implement ‘Sox lite’.
update process, often within an Risk professionals may protest that
automated toolset. At this point the this is finance work, but that does
exercise is almost entirely without not fly if the CRO or Head of Risk
value; in fact it makes rather a mockery of what the assessment reports to the CFO. Besides, why duplicate the risk and internal
process should be trying to achieve. It is an abdication of risk audit assessment process if it can just be extended to work for
management responsibility, or at the very least, outsourcing to Sarbanes-Oxley as well? So operational risk headcounts have
much less qualified people to do it. spawned, but no one in the business is seeing the commercial
Furthermore, if a firm has an adequately resourced internal benefit of this new industry, because its focus has become too
audit department, self-assessment is often a substantial compliance driven.
duplication of what auditors are supposed to be doing We have talked about the flawed process of self-assessment,
independently. There is nothing more depressing for the risk but Sarbanes-Oxley is self-assessment on steroids. A colleague
department than being the reconciliation agent between business once shared an article early on when Sarbanes-Oxley was first
self-assessments and internal audit’s risk assessment. Given the being introduced, that it was at heart a don’t ask, don’t tell regime.
problems of gaming outlined above, in my experience the internal The example given was the one about the father who tells his
audit assessment is usually much more valuable. Self-assessment teenage daughter to behave when she goes to a party, and when
in my view is largely debunked. she appears in the morning looking bedraggled, not having slept
If all risk assessment is independent, it is then simply a matter much and quite hung over, he asks her if she behaved. ‘Yes Daddy’
of which elements the risk department performs and which are she replies, because she cannot give any other response. Like so
conducted by internal audit. Each firm must work out this boundary many outputs of self-assessment, this makes a mockery of risk
for itself, but intuitively, risk will own the risk framework and risk management, and sadly reflects a giant leap backwards in the
quantification including risk indicator metrics, and audit will assess development of the discipline.
the design and effectiveness internal controls and conduct some
measure of substantive testing. Start Again from the Top Down
I know many banks that have beaten their chests over their
Sarbanes-Oxley and the Industry of Compliance operational risk implementation because they hired armies of self-
I was once asked at a conference how my firm was integrating its assessment teams and gleaned little or no commercial benefit from
operational risk and Sarbanes-Oxley programs. I was the last to it. If anything, they got quite the opposite — they wasted a lot of
answer on a panel of four; the other participants were all senior risk business time and focus, lost in the undergrowth, unable to see the
officers at major US banks. Each had given a drawn-out response forest for the trees. None of them dares to call a halt though. What
but I simply said that my firm was not US listed and did not have to CRO wants to give the impression that he or she is not serious
comply with Sarbanes-Oxley. I immediately received a somewhat about all risks, however small? And if the CRO stopped, would
unexpected rapturous applause. there not be pressure to start again after a loss? But the calling card
Operational risk professionals and certainly the vendor market is poor, has put risk in the space of audit, and seriously damaged
had initially welcomed the budgets that came with the new the brand. Op risk has lost so much credibility over the industry of
Sarbanes-Oxley rules. The enthusiasm was short-lived. Up to that control self-assessment.
point, operational risk, although perhaps thinly resourced, had Instead, CROs and their op risk heads should have been
enjoyed a relative period of invention: regulators were happy to ‘let religiously focused on the top down perspective, collecting
a thousand flowers bloom’ in development of the discipline. The objective data to support their scenarios, and drilling into the detail
general impetus of the profession had been to collect data and give as and when needed. That way they would have been in a position
objectivity to the assessment process. Implementing operational to keep their departments lean, high quality, motivated and
risk management was about achieving a commercial benefit from commercially oriented. Unfortunately, few have had the courage or
understanding and managing risk exposures better — it was not shown the leadership to follow this path on a consistent basis. Too
about regulatory compliance for its own sake. many op risk functions became the collection points for disparate
Unfortunately, the new budgets that arrived with Sarbanes- risk assessment cottage industries rather than the strategic,

T H E T R O U B L E D I N D U S T RY O F R I S K A S S E S S M E N T O C TO B E R 2 0 1 2 9
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

business-minded and advisory functions they should have been.

Therefore, my message to the CRO who is disillusioned with op ABOUT THE AUTHOR
risk implementation is that it is not too late to start again and
refocus the activity. Assess the big risks and hold capital for them, Jonathan Howitt has extensive experi-
but manage the rest of the risk implementation as an expected loss ence in risk management on the buy and
management program. Of course, processes will need to be sell sides. Most recently he held Group
mapped and risks understood regarding control dependencies for Risk and CRO roles at two major UK listed
major activities. Once that is done, it is fairly static information. investment firms with businesses in alter-
Risk then needs to be dynamic and data-driven to be useful for the native fund management, brokerage,
business. Invest in collecting risk information and synthesizing it, investment banking and private wealth
and keep looking for new angles on data. Do not stop reinventing management. He has also held senior
your KRI reports and loss analysis, and firm up the link with Finance, COO and Risk roles in New York, Tokyo and London
remediation. And when your assessments need updating, do not with Citibank, UBS and Dresdner Kleinwort, and during the
do this in a vacuum or bury it in an IT toolset. Engage the business early 2000’s was directly involved in industry discussions
in a positive and interactive way — use risk workshops, brainstorm with regulators for Basel II.  He is a member of PRMIA’s
the issues and demonstrate the valuable contribution of the risk Publications Committee and also the London Chapter
Steering Committee.
function as objective facilitator and trusted adviser; a catalyst not
a cost.

Three reasons to choose PRMIA for your in-house training needs

Independent Association  ✔ Training is delivered in classroom, online and by webcast. 

Customized Content  ✔ PRMIA is dedicated to providing resources, networking,

and thought leadership to help our members achieve the
Qualit highest standards from the cradle to the pinnacle
of their careers. PRMIA will guide you through these
tough economic times by providing you with the
educational and training opportunities needed to
strengthen your risk management knowledge and skills.
✔As a professional association, our global network
provides us access to trainers who are industry
leaders and subject matter experts.

E-mail [email protected] to schedule a conversation.

10 INTELLIGENT RISK VISIONS OF RISK

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
VISIONS OF RISK THOUGHT PIECES FROM PRMIA LEADERS

RISK AS AN AGENT OF CHANGE

WILLIAM MASON

R isk based supervision starts with the premise that not all firms are equally important to the economy and that a
supervisor can deliver most value through focusing its energies on the firms that are most significant and on the risks
that pose the greatest threat to financial stability and consumers. A risk based system should also provide a systematic
and structured means of assessing different types of risk, ensuring that idiosyncratic approaches to firm supervision are
avoided and potential risks are analyzed and mitigated using a common framework.

Over the past year, the Central Bank of Ireland has introduced more likely to help a supervisor identify and act upon the key
PRISM — a new risk-based framework and IT platform for risks at a regulated firm. We worked hard to defend the key
managing the supervision of the approximately 10,000 financial parts of that logic in the face of internal challenge throughout
firms for which we regulate. We have done this as part of our the design and build phases, while trying to accommodate
program to improve regulation in the wake of the 2008 financial others’ views. Having logical integrity was doubtless a pre-
crisis, and building upon lessons learned from specific domestic condition for subsequent successful operation of the
failings highlighted in various reports by Honohan, Nyberg and framework, and helped secure senior management team buy-
Regling and Watson. in but it did not, in and of itself, ensure acceptance from all
PRISM makes it easier for our supervisors to challenge the firms potential users. Front line staff, while they may have
they regulate, judge the risks they present and take action to appreciated the logical case we were making for change,
needed something more than that to encourage them to
mitigate those risks — securing meaningful change on behalf of
engage in the project.
consumers, citizens and the state. PRISM focuses attention on the
firms that have the highest potential impact on the economy,
■ Engagement — Rather than focusing on theory, we found that
making it materially less likely that they will fail in a disorderly
structured engagement with those who were required to
fashion. PRISM moves the Central Bank to a unified risk based
change their practices as a result of PRISM adoption yielded
system designed to make every euro we spend on supervision go
significant benefits:
as far as possible, a system which encourages supervisors to focus
on the issues which really count. • In the first instance, we recruited representatives from all the
For supervisors, adapting to the new approaches to supervision supervisory divisions to participate in the graphical user
embedded within PRISM has involved a lot of flexibility and a lot of interface (GUI) design process. This meant that we had the
change. As financial institutions adapt to the new regulatory and look and feel of all the system’s modules structured by
political realities, many are likely to be in need of improvements to practical users rather than professional programmers. It also
their practices and will be required to undertake significant change meant that we had, de facto, a cadre of high caliber change
agents within each division.
programs in order to meet the new higher expectations of the
Central Bank and the wider public. In the article I highlight some of • The benefits of visual stimuli to give future users an
the lessons we learned and some of the things we did which proved understanding of the positive aspects of change cannot be
successful in our work to renew supervision at the Central Bank: overestimated. PRISM is both a framework for supervision
and a bespoke IT system. We went to some trouble to show
■ Logic — Change is often used as a mantra — almost an end in an early incomplete system to future users and managers to
itself — by managers who fail to explain what it means and why allow them to see what it would look like. This made what we
it will make life better for the employees who are to experience were talking about much more real and, as future adopters
the change. Conscious of this, prior to designing and building saw that it was considerably more clever than an Excel
PRISM we studied regulatory frameworks, such as those used workbook, they began to believe that change might be
by Australia and Canada, to ensure that what we constructed positive, time-saving and empowering.
was both fit-for-purpose and reflected current regulatory best
• When training supervisors, we engaged with senior
practice. We then developed a theoretical framework drawing
supervisors and involved them in the delivery of parts of the
on that best practice, with input from senior staff in our own
training. In this way, we enabled them to demonstrate line
organization. Critical to the intellectual robustness of the
management buy-in to PRISM. We also invested heavily in
framework we were trying to construct was to obtain top-level
credible senior trainers with an in-depth knowledge of
buy-in. Coherent logic underlying the entire system has been
supervision who easily gained the respect of those they were
vital in ensuring that it is intellectually robust and therefore
training.
R I S K A S A N AG E N T O F C H A N G E O C TO B E R 2 0 1 2 11
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

• Having formally launched PRISM we held follow up training some of the ideas directly contradicted other ideas. We found
sessions that allowed us to support supervisors doing real using formal forums to manage the feedback essential to
work on the system for the first time. This helped bridge the ensure that the top tier of management in the Central Bank
gap between theoretical exercises and actually making the discussed the good aspects and weaker aspects of ideas, with
system work for supervisors for the first time. They went decisions clearly minuted. At such forums, the risk division had
away from these training sessions having started to populate to learn to bend with the wind and stretch to adapt the system
the system with risk analysis on their firms. to meet emerging client needs, while preserving key aspects of
its logic. The unswerving support of the Deputy Governor,
• Following the first release of the system we had structured
Financial Regulation was vital and I would observe that any
feedback meetings with key user groups to understand how
head of risk who agrees to implement a major framework
they were finding the system and what practical capabilities
change without the active support of his or her CEO, is taking
they were finding it lacked. By responding quickly once we
a “courageous step” as Sir Humphrey Appleby would have
had listened to feedback, we were able to incorporate several
put it.
of their suggestions for improvement into the second phase
of the build, delivering the improvements swiftly. Such
We certainly did not get everything right while designing and
responsiveness on our part helped to build trust.
delivering PRISM at the Central Bank. It was fascinating to observe
the necessity of and yet the limits of a good theory as a change
■ Support — Without appropriate support, we would have found
it impossible to deliver a high quality system to very tight agent and the vital importance of the many types of participatory
deadlines. We spent a lot of time and energy procuring high engagement we deployed with those who would be called upon to
quality external support. Our principal IT build consultants make the change a reality in our “front office." The importance of
committed themselves to giving us the same team for all four fully committed teams who understood and were prepared to
phases (one design stage and three stages of build) with the strive for the strategic benefits of the change was brought home to
same people co-located in our office throughout the project. me again and again as deadlines approached and people — risk,
This gave us a great deal of flexibility in responding to feedback supervisors and in-house IT — in a public service environment —
and achieving fast turnaround times. Further to that, we used worked very long hours to deliver their pieces of the jigsaw puzzle.
external consultants to act as a thought partner on aspects of Having consultants who were flexible enough to run with you
our regulatory thinking – providing impartial advice on the logic rather than reading to you out of their manuals on how to run was
and practicality of our design. Switching to internal support, I also important for a bespoke framework. Our software developers
was given a headcount rather than a team when I took up my could be pushed to achieve, but were also quite adept at indicating
post and was asked to design the new system. This was hard the limits to the possible rather than overpromising and under
work at first but it gave my deputy and me a lot of flexibility to delivering.
recruit staff from a broad range of backgrounds — people who In summary, logic, engagement, support and understanding
signed up because they wanted to make a difference and liked feedback are all vital when designing and implementing a major
the sound of the change project that PRISM had become. organizational change, as are, of course, a good appreciation of the
While this had occasional downsides in terms of my teams not art as well as the limits of the possible.
always having experienced line supervision, it ensured that
there was a lot of constructive challenge from those coming The views in this article are the author’s own and do not necessarily
into the risk division without supervisory experience. This represent the official views of the Central Bank of Ireland.
forced us to improve our standards of communication. I ended
up with a team of enthusiastic volunteers — no one was there
because they had been told they had to be. It also meant that
there were few people who were unduly attached to previous
approaches — an issue for some change projects. We were ABOUT THE AUTHOR
also blessed with a supportive COO and an IT department that
was intellectually interested in what we were doing and William Mason heads the Central Bank of
prepared to experiment with new approaches and put itself out Ireland's Risk Division. He led the design
and build of its new risk-based supervision
to meet our tight deadlines.
system - PRISM. Prior to joining the
Central Bank, he supervised a range of
■ Feedback — Managing feedback was very important
credit institutions, insurers and invest-
throughout the process. Lots of people had constructive ideas
ment firms at the UK Financial Services
as to how the new system should work. The issues we faced
Authority. He also authored regulatory improvement propos-
were, first, that not all the ideas could be practically als at the UK’s Better Regulation Task Force, including
accomplished in a design phase measured in weeks and a build 'Regulation — Less is More — a report to the Prime Minister’.
phase measured in a few short months and, secondly, that

12 INTELLIGENT RISK VISIONS OF RISK

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
VISIONS OF RISK SPONSOR ARTICLE

MODELLING CREDIT SPREADS FOR COUNTERPARTY RISK:

MEAN-REVERSION IS NOT NEEDED
IGNACIO RUIZ AND PIERO DEL BOCA

Abstract
When modelling credit spreads, there is some controversy in the market as to whether they are mean-reverting
or not. This is particularly important in the context of counterparty risk, at least for risk management and capi-
tal calculations, as those models need to backtest correctly and, hence, they need to follow the “real” measure,
as opposed to the “risk-neutral” one. This paper shows evidence that the credit spreads of individual corporate
names, by themselves, are not mean-reverting. Our results also suggest that a mean-reversion feature should be
implemented in the context of joint spread-default modelling, but not in a spread-only model.

Arguably, building a credit model for counterparty risk has some January 2005 to January 2012 for 23 corporate names covering a
important challenges. There are a number of well-developed and range of credit qualities. Figure 1 shows those time series.
sophisticated models for this asset class in the literature, but they
have mostly been developed in the context of derivatives pricing.
A pricing model is good when it provides (i) no arbitrage gaps,
(ii) good greeks that enable hedging strategies and, obviously,
(iii) reasonable prices. In the world of risk management and capital
calculation, a model is good when it backtests properly. That is, risk
management needs the distributions arising from the models to be
as close as possible to the actual distributions seen in the market.
For that reason, while derivatives pricing models typically use the
“risk-neutral” measure, risk management and capital calculation
models tend to use the “real" measure.
In particular, in the credit asset class, one of the stylised
facts that credit spread models often mimic is mean-reversion.
However, there is some controversy as to whether the credit
spreads are mean-reverting or not. In early studies of credit
spreads, they were considered mean-reverting, but this was
not the case in later studies.[1, 2].
This paper tackles that question: are credit spreads mean- Figure 1: Times series of 5-year credit spreads for a number of
reverting? names, from January 2005 to January 2012. Source: Credit Suisse.

Assessing Mean-Reversion In order to assess the validity of the measured mean-reverting

In order to answer that question, we calibrate a number of parameters, the authors did two tests:
times series of credit spreads to both an Ornstein-Uhlenbeck (OU)
process 1. A Null Hypothesis Test.

d xt = θ o u ( μ o u – x t )d t+ σ o u dWt
2. A GBM Compatibility Test.

and a Black-Karasinski (BK) process Null Hypothesis Test

d log ( x t ) = θ bk ( log ( μ bk ) – log ( x t ))d t+ σ bk dWt .

Figures 2 and 3 show the mean-reversion speed (θ ) calibrated to
those time series, both for an OU and a BK process, together with
error bars1. The graphs also include the mean-reversion speed
The methodology used for the calibration of these models was obtained from the VIX time series, for comparison, as the VIX time
an Ordinary Least Squares (OLS) regression. series is expected to show mean-reversion towards a VIX value of
The authors used time series data of 5-year spreads from around 20%2.

1. Error bars indicate the 99% confidence interval assuming a normal distribution with the empirical standard deviation.
2. See VIX time series in Appendix.

M O D E L L I N G C R E D I T S P R E A D S F O R CO U N T E R PA R T Y R I S K O C TO B E R 2 0 1 2 13
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The reader can see in those graphs how the zero value is
included within the error bars in most cases.
Let’s say that the null hypothesis is that “the mean
reversion parameter speed θ is zero for corporate credit
spreads”. The data shown in Figures 2 and 3 illustrates
that this null hypothesis should not be rejected3.

GBM Compatibility Test

The authors did a further test which, in their view,
provides a more intuitive insight into the problem.
They generated 10,000 simulated paths using a
Geometric Brownian Motion (GBM)4 process and, then,
they measured the mean-reversion speed with the same
Figure 2: Mean-reversion Speed of 5-year credit spreads for an OU process. procedure as used for the real data. If those GBM paths
were in nitely long, then the mean-reversion speed
calibrated out of them should be zero, but the GBM paths
were identical in length and granularity to the credit
spread time series (7 years, daily data) and, so, the
calibrating algorithm will deliver non-zero values for θ
from each of the GBM paths, even when we know that
they are by construction non mean-reverting. From those
10,000 θ ’s, the authors constructed statistics regarding
the values of the mean-reversion speed compatible a non
mean-reverting process sample (the GBM process in
this case).
The distribution of those “artificial” values of θ are
shown in Figures 4 and 5 for an OU and a BK process
respectively. They are shown together with the values
obtained from the time series of the credit spreads
Figure 3: Mean-reversion Speed of 5-year credit spreads for an BK process.
(green dots) and, for comparison purposes, with the
value obtained from the time series of the VIX index
(red dot).
Those two figures show how the mean-reversion
speed obtained from the credit spread time series are
completely compatible with a GBM non mean-reverting
process. This is in contrast with the θ obtained from VIX,
which is well outside the values compatible with
a GBM process5.

Credit Spreads Are Not Mean-Reverting, at Least in the

“Classical" Way
The authors have calibrated the mean-reversion speed of
an OU and a BK process to the time series of 5-year
credit spreads of 23 different corporate names covering a
range of credit standings, and they have shown that (i)
Figure 4: Distribution of mean-reversion speed calibrated to an OU process
the error of those values makes it unreasonable to reject
in 10,000 artificially generated GBM paths (blue), together with the mean-
reversion speed values calibrated from the credit spread time series (green)
the null hypothesis6 and (ii) those values seem to be
and the VIX time series (red). compatible with a GBM process, which is non mean-
reverting.

3. Strictly speaking, we should expect that in a few out of the 23 measurements, the zero value for θ should be outside of the 99% confidence interval even when
the null hypothesis is not rejected. However, the authors are not being specific about how many is “a few" as there is uncertainty regarding the actual distribution of
the error. They have assumed that the latter is normally distributed, but that might not be the case. In any case, in spite of this lack of strict precision, the authors
think that it will be unreasonable to reject the null hypothesis given the data obtained.
4. With a volatility of 100%, which is a typical volatility observed in the time series of the credit spread time series.

14 INTELLIGENT RISK VISIONS OF RISK

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
In their view, this clearly indicates that credit spreads of this mean-reversion must be a joint spread-default process so that,
individual names should not be modeled with a mean-reverting loosely speaking, the source of the mean-reversion is not in the
process in a “classic” way. By “classic” it is meant via an OU, BK or spread evolution itself, but rather in the dependency between
a similar process. spread evolution and default events.
However, having said that, it appears evident from market
observation that credit spreads display some sort of mean- Appendix
reversion feature subject to survival. This comes from the fact that,
should a corporation have a credit spread of, say, 8,000 bps, we
would expect it to either default soon or, if it survives in the long
run, drift towards a much lower credit spread. So, mean-reversion
should be captured by credit spreads subject to survival, not by
credit spreads by themselves. As a result, a credit model capturing

Figure 6: VIX index time series. Source: Bloomberg.

References
[1] H. Bierens, J. zhi Huang, and W. Kong, An econometric model of credit
spreads with rebalancing, arch and jump effects. Working Paper, 2003.

[2] F. A. Longstaff and E. S. Schwartz, Valuing credit derivatives, Journal

Figure 5: Distribution of mean-reversion speed calibrated to an BK of Fixed Income, 7 (1995), pp. 6–12.
process in 10,000 artificially generated GBM paths (blue), together The authors would like to thank Seung Yang and Chris Kenyon for
with the mean-reversion speed values calibrated from the credit interesting and useful remarks on this paper. The content of this article
spreads time series (green) and the VIX time series (red). reflects the views of the author, it does not represent his employer’s views or
opinions in any way.

5. Again, strictly speaking, this exercise should be done per measured θ . Ideally, the following should be done: (i) measure θ for each name and model (as it
has been done) as well as the volatility σ , (ii) generate 10,000 paths for each one of the 23 time series and model (OU and BK) with θ = 0 and with a volatility
of σ and (iii) calculate a p-value of each θ from which the validity of each mean-reversion speed can be assessed. Instead, the authors have taken an average
volatility of 100%, generated 10,000 GBM paths (as the credit spreads cannot be negative) and use these statistics to validate all values of θ . However, the con-
clusion using this shortcut are so clear that the authors think that the strict analysis explained in this footnote is not strictly necessary.

6. Null hypothesis: “the mean reversion parameter speed θ is zero for corporate credit spreads”.

ABOUT THE AUTHORS

Ignacio Ruiz, founding director of iRuiz Piero Del Boca is a quantitative risk analyst
Consulting ltd, London, established himself at Credit Suisse. Before that he was a quan-
as a contractor and independent consultant titative fixed income analyst at the asset
in Quantitative Risk and CVA in 2010. management division of Eurizon Capital at
Before that he was the head quant for Milan. He holds an MSc in theoretical
counterparty credit risk, exposure meas- physics from Università degli Studi di
urement, at Credit Suisse and the head of Milano, Italy, and an MSc in financial math-
equity risk methodology at BNP Paribas. He ematics from Warwick Business School,
holds a Ph.D. degree in nano-physics from Cambridge University, UK. [email protected]. Piero Del Boc is not com-
UK, and an advanced degree in theoretical physics from mercially related to iRuiz Consulting Ltd.
Complutense University, Spain. www.iruizconsulting.com
[email protected]

M O D E L L I N G C R E D I T S P R E A D S F O R CO U N T E R PA R T Y R I S K O C TO B E R 2 0 1 2 15
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
VISIONS OF RISK SPONSOR ARTICLE

REVERSE STRESS TESTING FROM A MACROECONOMIC VIEWPOINT:

Quantitative Challenges & Solutions for its Practical Implementation
JUAN M. LICARI AND JOSÉ SUÁREZ-LLEDÓ

T he Financial Services Authority defines reverse stress tests as “tests that require a firm to assess scenarios and circum-
stances that would render its business model unviable, thereby identifying potential business vulnerabilities. Reverse stress-
testing starts from an outcome of business failure and identifies circumstances where this might occur. This is different to gen-
eral stress and scenario testing which tests for outcomes arising from changes in circumstances.”
At its core, reverse stress testing (RST) proposes to “invert” the stress testing process f : X → Y will help us ensure that (at least

mapping f -1 : Y → X . Applications of results such as “Inverse

standard process; starting now from an “outcome” (business failure) locally) one can “invert” the process and obtain a reliable RST
with the aim of finding potential states-of-nature (scenarios)
consistent with such outcome. When implementing RST in a Function” and “Implicit Function” theorems are applied to

~
X ={ x~ ∈ X ⊆ R̋ ˄ ~ x ∈ f -1 (y 0)} . Under some “regularity”
quantitative way, multiplicity emerges as a challenge. The same understand the shape of the solution-set

~
conditions for f , the set X will locally map X → Y (stress
outcome (for example, high expected losses) could materialize

testing) and Y → X (reverse stress testing) in a one-to-one

under multiple combinations of risk factors — such as probability
of default (PD), exposure at default (EAD), and loss given default
(LGD) — and under alternative macroeconomic scenarios. smooth fashion (overcoming type-2 multiplicity).
Multiplicity should definitely pose a concern, as the reverse
engineering exercise can end-up identifying only a subset of 2 — Multiplicity from a Practical Viewpoint
scenarios that are consistent with the starting assumption. There is Handling Type-1 Multiplicity:
risk that such scenarios that were not found could be of more Factor & Principal Components Analysis
relevance — and severity — compared with the ones that were By leveraging on the strong correlation of macroeconomic series, a
identified by the process. modeler can concentrate on a smaller set of instruments (factors)
that can still replicate most of the variability of the whole sample.
1 — Multiplicity from a Mathematical Viewpoint The main advantage is the reduction of the “scenario-space” (to
Type-1 Multiplicity: Indeterminacy avoid indeterminacy). The challenge, however, could be the lack of
Most RST frameworks are faced with the task of matching a large interpretation for the factors.
number of risk and macroeconomic variables with a limited set of We carried a case study to link the top factors to specific
assumptions; i.e., # variables > # equations. Indeterminacy takes macroeconomic series, ensuring that intuition is kept on the nature
the form of a continuum of solutions (scenarios) whose of the factors while reducing the chances of indeterminacy. Below
mathematical properties will help the modeler identify avenues to are the findings for the UK; similar results were found for US and
close the extra degrees of freedom. This indeterminacy needs to be Germany.
dealt with before any further attempt is made to successfully
reverse engineer the process. Solutions will require (a) additional
ad-hoc assumptions on some parameters (expert-judgment,
market-wide assumptions or values in line with regulatory

findings; e.g., LGD=f(PD) . The task is to close the “degree of

guidelines), and/or (b) additional equations based on empirical

indeterminacy” to zero and end-up with as many equations

as unknowns.

Type-2 Multiplicity: Inverse Mapping

Even after closing the gap between equations and unknowns a new
challenge emerges when trying to reverse engineer a process — the
inverse of a function may not behave as a function. Consider a
stressed value of the risk factors, say x 0 , that is mapped to a vector
of outcomes, say y 0 = f( x 0 ) . Mapping y 0 back (i.e., reverse
engineering the process) could give us a value of x that is different
from x 0 : there may exist another vector x 1 that is consistent with
the same outcome y 0 = f ( x 1 ) . Specific characteristics of the

16 INTELLIGENT RISK VISIONS OF RISK

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

The top five UK factors explain over 90% of the variability of the replicated by these five factors. But what do these factors
whole sample, with factors 1 and 2 explaining 50% and 20%, represent? Figures I to VI show the correlation of the factors to
respectively. This is encouraging news from a RST angle; almost all specific economic indicators.
of the information embedded in the UK economic cycle can be

R E V E R S E S T R E S S T E S T I N G F R O M A M AC R O E CO N O M I C V I E W P O I N T O C TO B E R 2 0 1 2 17
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

Consider a linear model: Y 1 = β X 1 + ε 1 , where Y 1 ∈ R M

represents a vector of risk variables or targets and X 1 ∈ R N
The dimension of the “scenario-space” has been dramatically
reduced while keeping an interpretation of the factors: (1) core real

the form Y 1+1 = βˆ X 1+1 , with β̂ representing the estimated

business cycle fluctuations, (2) labor market, (3) monetary cycle contains all the model drivers. The (one-step-ahead) forecast takes
F
(inflation and rates), (4) money supply (complementing factor 3),
and (5) UK housing market. With this one-to-one match between parameters.

assumption for the outcome, say Y 1+1 , and find consistent values
macro variables and factors, the modeler responsible for RST has a
s
Suppose now that the RST mandate is to start with an

for X 1+1 . This implies ( βˆ ́βˆ ) -1 βˆ ́Y 1+1 =X 1+1. If N > M , we are

better chance of succeeding, due to the lower number of variables
s

back into type-1 multiplicity; so N =M is necessary to continue

to be matched with outcomes/targets.

2.2 — Handling Type-2 Multiplicity: The Case of Linear Models working on the RST process. With an equal number of targets and

to be verified is the full rank of βˆ ́βˆ (or a non-zero determinant).

Risk modelers usually apply non-linear transformations to risk control variables, the necessary and sufficient condition that needs

Ensuring that βˆ ́βˆ is “invertible” will avoid the presence of type-2

variables (e.g., logistic or logarithmic mappings) and then model

multiplicity and the scenarios in X 1+1 will be uniquely linked to the

these transformed series in a linear fashion against macro-

outcome Y 1+1 .
economic and other risk drivers. Properties of the coefficient
matrices of these linear systems will determine whether the
process can be “inverted”. The simplest 1-dimensional linear model With non-linear transformations of the original risk variables
will require a non-zero estimated coefficient. When dealing with that are strictly monotone (as it is the case for logarithmic and
higher order systems, the non-zero condition translates to the logistic mappings), the modeler is able to avoid type-2 multiplicity
determinant of a specific matrix. and the RST exercise can be carried forward.

ABOUT THE AUTHORS

Dr. Juan M. Licari is a Senior Director at Dr. Jose Suarez-Lledo is an associate direc-
Moody’s Analytics and the head of the tor at Moody’s Analytics based in London.
Credit Analytics team for Europe, the As part of the Credit Analytics team he
Middle East, and Africa. Dr. Licari’s team designs retail and corporate credit models
provides consulting support to major as well as macro-econometric models for
industry players, builds econometric tools key economic and financial variables.
to model credit phenomena, and imple- Before joining Moody’s Analytics, Jose held
ments several stress-testing platforms to quantify portfolio risk a research position at the Universidad Autonoma de Barcelona,
exposure. His team is an industry leader in developing and imple- where he developed models for illiquid financial markets and the
menting credit solutions that explicitly connect credit data to the dynamics of asset prices and credit. Jose holds a Ph.D. and an
underlying economic cycle, allowing portfolio managers to plan MA in Economics from the University of Pennsylvania.
for alternative macroeconomic scenarios.
Juan is actively involved in communicating the team’s
research and methodologies to the market. He often speaks at
credit events and economic conferences worldwide. Dr. Licari
holds a Ph.D. and an MA in economics from the University of
Pennsylvania and graduated summa cum laude from the
National University of Cordoba in Argentina.

ABOUT THE SPONSOR Moody’s Analytics helps capital markets and risk management professionals
worldwide respond to an evolving marketplace with confidence. The company
offers unique tools and best practices for measuring and managing risk through
expertise and experience in credit analysis, economic research and financial risk
management. By providing leading-edge software, advisory services, and research,
including the proprietary analysis of Moody’s Investors Service, Moody’s Analytics
integrates and customizes its offerings to address specific business challenges.
For more information please visit: www.moodysanalytics.com/RSTmacro

18 INTELLIGENT RISK M O O DY A N A LY T I C S S P O N S O R A R T I C L E
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CHAPTER REPORT

PRMIA PHILIPPINES CHAPTER

T he Philippines Department of Tourism recently reintroduced the Filipino FUN through its new tagline — IT’S MORE
FUN IN THE PHILIPPINES. This concept is captured by not only the nation’s beautiful scenery, exotic attractions
and daily life in general, but also by the lively mood, bright smiles and limitless humor of its citizens. Filipino FUN gives
the whole world a peek into the resilience and positivity of the Filipino spirit, which remains undaunted by tragedy, polit-
ical unrest and day-to-day challenges.

The PRMIA Philippine Chapter began in 2003 as an

organization of 23 members focused on financial serv-
ices risk management, and has evolved into a diverse
organization whose 2012 membership total is over
700. The chapter is now a melting pot of risk profes-
sionals representing academia, government, financial
and non-financial services — a multi-discipline
exchange of ideas and perspectives.
After a period of inactivity surrounding changes in
leadership, the chapter has rebounded significantly
over the last twelve months. Recently sponsored
events have featured speakers from the Central Bank
of the Philippines (BSP), risk management profession-
als and industry experts. The chapter’s event and sem-
inar offerings have changed with the times, and now
feature thought leadership in traditional risk manage-
ment areas, as well as emerging risk disciplines and
contemporary issues from regulators, members and invited friends. Indeed, Filipino FUN translates not only to enjoyable memories
Recent topics include an assessment of the Internal Capital of vacation and collective festivities, but also to everyday personal
Adequacy Assessment Process (ICAAP), operational risk tools and and professional life—even in the managing of risk.
techniques, stress testing, and the US Foreign Account Tax
Compliance Act (FATCA).
The members of the Chapter’s Steering Committee and
The chapter takes a practical approach, aiming to transcend
key volunteers are:
beyond mere theories and principles. While the discussion and
debate among participants is often passionate, everyone enjoys a • Rose Javier, co-Regional Director
few laughs with colleagues and new friends, and goes home with a • Raffy Dayrit, co-Regional Director
smile, as well as an enhanced outlook on managing risk. The steer- • Ray Vergara, Treasurer
ing committee (SC), always the gracious and hospitable host, • Derrick Nicdao, Secretary
ensures that all attendees leave with a satiated mind, as well as • Roy Consulta, SC member
stomach. Participants have takeaway insights that they can then
• Alvin Dave Pusing, Events Committee chair
translate into action items within their respective organizations.
The chapter simply could not function without its formidable • Click San Juan, Events Committee member
backbone of three committees — steering, events and membership • Michelle Fajardo-Tegonciang, Events Committee member
— whose volunteers plan and execute chapter events from start to • Vincent Espiritu, Membership Committee chair
finish. With varied backgrounds, different roles and personalities, • Kristine Po, Membership Committee member
each committee meeting involves the exchange of ideas (both sen-
sible and crazy), hearty laughter and pizza. Consistent with the To learn more about Philippines Chapter of PRMIA, please visit
Filipino virtue of “bayanihan” (spirit of communal effort), the chap- www.prmia.org/Chapter_Pages/philippines or email us at:
ter has formed partnerships with academic institutions, local and [email protected].
international experts, government agencies and industry entities,
to deliver only the best to the chapter participants.

PRMIA PHILIPPINES O C TO B E R 2 0 1 2 19
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
WHAT’S ON THE WEB EXPERT GUIDES TO THE BEST LINKS

RISK MANAGEMENT AND GOVERNANCE OF FINANCIAL INSTITUTIONS

ANNA LINGEL

H aving a robust risk management function and governance system is seen as crucial after the global financial crisis.
Such factors as chief risk officer qualifications and status, executive compensation and board composition have
increasingly become a focus of regulatory bodies and researchers. While our research in the field of corporate gover-
nance regulation has drawn from many sources, we have found several websites that have been particularly important.
I have listed them below in hopes that readers of Intelligent Risk will find them useful as well.

THE NATIONAL BUREAU OF ECONOMIC RESEARCH – THE BASEL COMMITTEE ON BANKING SUPERVISION
RISKS OF FINANCIAL INSTITUTIONS http://www.bis.org/bcbs
http://www.nber.org/workinggroups/fr/fr.html
Rating: ★★★★
Rating: ★★★★★
This website is a great starting point for exploration of the topic of
The NBER is a private, nonprofit research organization. One goal of regulations and governance. A good example is “Principles for
the NBER's Working Group on the Risks of Financial Institutions is enhancing corporate governance” which provides the recom-
to improve the understanding of risks in financial markets and to mendations of the Basel Committee at
investigate how regulations affect these risks. The working group’s http://www.bis.org/publ/bcbs176.pdf. This 2010 document
website provides several research papers. Particularly interesting is summarizes the lessons learned from the financial crisis of
“Stronger Risk Controls, Lower Risk: Evidence from U.S. Bank 2007-08.
Holding Companies,” by Andrew Ellul and Vijay Yerramilli, which Other interesting papers from the Basel Committee include
analyzes the influence of strong and independent risk management “External audit quality and banking supervision” which describes
functions on enterprise-wide risk: the roles of supervisors and external auditors:
http://www.nber.org/papers/w16178 http://www.bis.org/publ/bcbs146.pdf and
Also see the research paper “Yesterday’s Heroes: “Core principles of effective banking supervision”
Compensation and Creative Risk-Taking” by Ing-Haw Cheng, http://www.bis.org/publ/bcbs213.pdf
Harrison Hong and Jose A. Scheinkman, which explores the
relationship between compensation and risk-taking among finance
firms: http://www.nber.org/papers/w16176 THE HARVARD LAW SCHOOL FORUM ON CORPORATE
GOVERNANCE AND FINANCIAL REGULATION
http://blogs.law.harvard.edu/corpgov/
RESEARCH AT THE WORLD BANK
Rating: ★★★★
http://bit.ly/53q9td
This Forum provides updates on working papers, seminars, speak-
Rating: ★★★★★
ers, and other activities in the field of corporate governance and
Three researches at the World Bank, James R. Barth, Gerard Caprio, financial regulation. One example, “Toward Effective Governance
Jr. and Ross Levine, examine a cross-country survey on bank of Financial Institutions,” features Lord Adair Turner, Chairman of
regulation in the years 2000, 2003 and 2007. Since regulatory the United Kingdom Financial Services Authority, discussing the
certainly plays such an important role in risk management, the features of an effective governance structure and offering his
results of these surveys are fascinating. recommendations: http://hvrd.me/Jn8sDk
In addition to providing raw survey data, the researchers have
published several papers analyzing the influence of regulation on
bank performance and how this has changed over the years
including: “Bank Regulation and Supervision: What works best”
http://bit.ly/RfmXvv and “Bank Regulations Are Changing: For
Better or Worse” http://siteresources.worldbank.org/INTRES/
Resources/ForBetter_or_Worse_final.pdf

20 INTELLIGENT RISK W H AT ’ S O N T H E W E B
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

THE INSTITUTE OF INTERNATIONAL FINANCE THE EUROPEAN CORPORATE GOVERNANCE INSTITUTE

http://www.iif.com/about/ http://www.ecgi.org

Rating: ★★★★ Rating: ★★★★

The website of the Institute of International Finance (IIF) offers The non-profit association ECGI provides research and advice on
insights on its approach toward the development of best practices the formulation of corporate governance policy. The ECGI publishes
and standards for the financial industry. The subcategory a Research Newsletter that gives an overview of the hot topics in
“Regulatory Affairs” discusses current developments in regulation corporate governance as well as the current research:
and their impact on banks. Many useful gems can be found here, http://www.ecgi.org/research/index.php
including the paper “Implementing Robust Risk Appetite
Frameworks to Strengthen Financial Institutions,” which gives Two series, the Working Paper Series in Finance and the Working
excellent practical guidance for those seeking to determine a risk Paper Series in Law include downloadable research papers from
appetite framework: several academics. One paper of interest is entitled “Why Did
http://www.iif.com/regulatory/article+968.php Some Banks Perform Better during the Credit Crisis? A Cross-
Country Study of the Impact of Governance and Regulation,” and
“Compensation Reform in Wholesale Banking 2011” was produced was written by Andrea Beltratti and René Stulz:
jointly by IIF and Oliver Wyman and provides the third annual http://www.ecgi.org/wp/wp_id.php?id=386
survey of its type:
http://www.iif.com/regulatory/article+1026.php The paper investigates whether bank performance is related to
bank-level governance, country-level governance, country-level
Another interesting publication is a survey that the IIF conducted in regulation, and bank balance sheet and profitability characteristics.
cooperation with Ernst and Young, “Progress in Financial Risk
Management” http://bit.ly/MgdxMd
THE SENIOR SUPERVISORS GROUP (SSG)
Additionally, the IIF publishes the Global Regulatory Update, which
Rating: ★★★
IIF members may access free:
http://www.iif.com/regulatory/gru/ The SSG is made up of senior representatives of several superviso-
ry authorities. Although the SSG does not have a specific website,
several regulatory agencies publish its works. One interesting arti-
cle, “Observations on Risk Management Practices during the
Recent Market Turbulence,” can be found at: http://bit.ly/lEJlm

Also of interest is the paper ”Observations on Developments in

Risk Appetite Frameworks and IT Infrastructure,” at:
http://bit.ly/ik9Qlx

ABOUT THE AUTHOR

Anna Lingel is a student at the University of Augsburg/Technische Universität Münich, currently working toward a
Master of Science degree in Finance and Information Management. As part of her degree, Anna is working on a
research project in risk management, with Elizabeth Sheedy, Associate Professor, Macquarie University Applied
Finance Centre in Sydney. For more information, please contact Anna at [email protected].

R I S K M A N AG E M E N T A N D G OV E R N A N C E O F F I N A N C I A L I N S T I T U T I O N S O C TO B E R 2 0 1 2 21
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
LEARNING OPPORTUNITIES

In the current environment risk education is not just a choice, it is a necessity.

S ince the global recession began in 2008 the demand for risk management training has dramatically increased
at all levels. In response, our training is evolving in line with member needs. PRMIA recognizes the diversity in
this renewed demand and has responded by providing a library of risk education tools, delivered in brief via online
and web-based training solutions, as well as through live classroom and customized in-house training. All plat-
forms are created and delivered by leading industry experts.

ONLINE SERVICES
Available anytime, anywhere in the
world with an internet connection.

WEBINARS
Most of our webinars are free to Sustaining Members. They are
the perfect option for professionals that desire flexibility, are
cost-conscious, and do not want to sacrifice quality. PRMIA
webinars bring the international thought leaders in risk
management live to your screen. Ask them a question and CLASSROOM TRAINING
participate in questions to the audience. Or, being recorded, you
Intensive and Comprehensive
may prefer a time and location that is more convenient than the
broadcast slot. Already a global market leader in risk
CUSTOMIZED COURSES
management webinars, we are expanding our schedule even
Customized courses are held in-house or at specialized training
further during 2012. Watch your e-mail and check the website
venues. One-to-one consultation with our specialist training
for updated schedules on http://bit.ly/PRMIAWebinars
professionals ensures that the learning experience is tailor-
made to your requirements. Our goal is to provide training that
PROFESSIONAL DEVELOPMENT
is flexible and sensitive to delegates’ needs, knowledge and
PRMIA offers over 700 online professional development
background. Enquiries: [email protected]
courses, all of which can be customized to your personal or
corporate needs. Delivered individually or as a corporate
OPEN ENROLLMENT COURSES
package, online training is extremely cost-effective, with most
Open enrollment courses meet the needs of members who
individual courses priced at only US $25. Special pricing is
prefer to interact and network with other risk professionals
available for corporate licensing of any online course or course
while receiving a more rigorous training experience. All PRMIA
combination. See http://bit.ly/PRMIAOnlineTraining
courses are taught by risk management industry practitioners and
university faculty, offering a unique blend of teaching. We have
EXAMINATION PREPARATION
several classroom courses scheduled over the next few months
PRMIA offers access to multiple resources to assist candidates
and our schedule will continue to develop throughout 2012 as
in the exam preparation process. These resources include
we receive feedback and guidance from members and leaders.
printed publications, online training, webinars, classroom
See next page for current classes or click here for a list of
training and DVDs. A full list of online exam preparation material
upcoming courses.
is on http://bit.ly/PRMIAExamPrep

PRMIA OFFERS WEEKLY THOUGHT LEADERSHIP WEBINARS. TO REGISTER GO TO www.primia.org/webinars

“This short course on risk management (Complete Course in Risk Management) crams more into its 20-
week span than many other certificate or even degree level courses. The professors are excellent and the
material ensures a solid foundation in the subject. I would unhesitatingly recommend this course.”
Jay Namputhiripad, Director, Risk Management, Federal Home Loan Banks Office of Finance
28 W H AT ’ S O N T H E W E B
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
OPEN ENROLLMENT COURSES

A PRACTICAL APPROACH TO implement a more robust through-the-cycle funding model, is

OPERATIONAL RISK MANAGEMENT at the heart of bank risk management objectives today.
A two-day course led by Philippa Girling Mastering the principles of liquidity risk is imperative for all
Thursday and Friday, November 8–9, 2012 | Chicago bank executives. For more information go to
This course will equip you with the vital elements of an effec- http://prmia.org/events/view_events.php?eventID=T5204
tive operational risk framework, from loss data to risk and
control self-assessment, and from key risk indicators to sce- FUNDS TRANSFER PRICING AND ASSET
nario analysis. You will learn where you should focus your LIABILITY MANAGEMENT
efforts to ensure you meet your business and regulatory A one-day course led by Moorad Choudhry
requirements using a practical approach. You will develop Friday, March 22, 2013 | London
effective reporting strategies and will gain an understanding Internal funds pricing, of funds transfer pricing (FTP) in a
of the basics of capital calculation. The course also provides bank is an important ingredient in liquidity risk management.
you with proven generic templates to kick start your program. Treasury operating models need to ensure that the FTP
This course will also touch on case studies of operational risk mechanism is applied in a way that drives accurate pricing
in recent headline events, including MF Global and JP Morgan and returns analysis. A major concern for bank regulators in
Chase. For more information go to the post-crash era is that FTP, which makes it imperative that
http://prmia.org/events/view_events.php?eventID=T5084 the FTP regime in any bank be effective and fit for purpose.
For more information go to
COUNTERPARTY CREDIT RISK: THE IMPACT OF CVA, http://prmia.org/events/view_events.php?eventID=T5205
BASEL III, FUNDING AND CENTRAL CLEARING
A two-day course led by Jon Gregory A Complete Course in Risk Management
Monday and Tuesday, November 26–27, 2012 | London A five-day course offered jointly by PRMIA & Kellogg School
Counterparty credit risk and CVA have become key concepts of Management, Zell Center for Risk Research
since the outset of the global financial crisis. In addition, July 15–19, 2013 | Chicago
funding has become a key concern for assessing trading Effectively practicing risk management requires a broad and
costs. Regulation means that counterparty risk quantification solid foundation of knowledge. This intensive program, led by
and management will be key challenges over the coming the faculty of one of the world’s top business schools, pro-
years for banks and other financial institutions. The large vides delegates with just such a foundation. Beginning with
move towards centralized clearing for many types of OTC the essential elements of finance, risk measurement, markets
derivatives will also be a key dynamic defining structure of and financial instruments, students are then introduced to
financial markets. For more information go to the best practices of market, credit and operational risk man-
http://prmia.org/events/view_events.php?eventID=T5090 agement. The final sessions focus on integrating the knowl-
edge gained into a capital allocation framework and study of
ADVANCED LIQUIDITY RISK governance best practices. For more information go to
A one-day course led by Moorad Choudhry http://prmia.org/events/view_events.php?eventID=T5174
Friday, February 22, 2013 | London
Effective liquidity risk management is the key challenge for
For more information visit www.prmia.org
banks in the post-crisis Basel III era. The need to address reg-
or contact [email protected].
ulatory requirements on funding and liquidity, as well as
LEARNING OPPORTUNITIES O C TO B E R 2 0 1 2 23
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ACADEMIC PARTNER PROFILE SPOTLIGHT ON PRMIA’S UNIVERSITY NETWORK

KENT BUSINESS SCHOOL, UNIVERSITY OF KENT

RADU TUNARU

K ent Business School (KBS) at the University of Kent in Canterbury, UK, launched in 1988, has built a strong portfolio
of undergraduate, postgraduate and PhD degree programs, as well as non-degree ‘part time’ executive education
for both individuals and companies. These cover a wide range of activities, including risk analysis of financial products,
derivatives pricing, econometrics of financial markets and quantitative finance for capital markets. The 25 years of pro-
fessional and academic excellence have shaped KBS into a true intellectual hub for business students,
academics, local government and the private industry. As a result, KBS is now among the top 25 business schools in the
United Kingdom and enjoys an established reputation for research collaborations with the business industry.

The one-year Master’s in Financial Markets, currently seeking published in various top finance journals including Journal of
accreditation from PRMIA, has been designed to cover advanced Banking and Finance; Journal of Empirical Finance; Journal of
and core topics required by finance industry employers worldwide, Futures Markets; Quantitative Finance; International Journal of
including risk management, derivative products, corporate finance, Theoretical and Applied Finance; Journal of Portfolio Management;
market and credit risk analysis, econometric modeling of financial European Journal of Finance; Journal of Investing; Review of
time-series, financial engineering and fixed income. In addition to Quantitative Finance and Accounting; Journal of International
the standard lecture and workshop/ seminar training, students Financial Markets; Institutions and Money; Applied Financial
enrolled in this Master’s program dedicate the summer term to a Economics and European Financial Management.
15,000 word dissertation on a financial topic of their choice, under The status of PRMIA academic partner — which reflects the
personalized guidance from KBS academia, thus ensuring that core intrinsic value of our staff and programs in finance — places KBS
and advanced knowledge is grounded and students are fully alongside top providers of financial education worldwide.
prepared for the next stage in their careers. KBS also offers an MSc Furthermore, the accreditation by PRMIA, which we currently seek
in Financial Services in Banking. Core modules for this program for our MSc in Financial Markets, represents another notable
include financial regulation, financial institutions, commercial and milestone in the integration of this program in the international
investment banking and corporate finance. KBS students have gone elite of Master’s programs in Finance. In addition, the finance group
on into careers in the financial industry as derivatives traders, at KBS is currently exploring academic links in postgraduate
corporate finance and risk management analysts, start-up finance finance, taught and researched with reputable institutions from
specialists, business financial managers, and research associates, Brazil, Hong Kong and Switzerland.
to mention a few of the opportunities opened after graduating with Recently, we established relationships with financial services
a KBS MSc. For more information on our programs, visit: companies interested in specific projects and joint events. A
http://www.kent.ac.uk/kbs/programmes/masters/specialist notable example is the UK accountancy firm Crowe Clark Whitehill
finance.html LLP — part of Crowe Horwath International, the ninth-largest
Members of the academic staff in finance at Kent Business association of professional audit and advisory services firms in the
School conduct research in diverse areas such as credit and market world — where some of our best students will enhance their
risk modeling, hedging, financial markets integration, theoretical practical knowledge. For more information please visit our website
financial econometrics, and model risk. Their work has been at: http://www.kent.ac.uk/kbs

ABOUT THE AUTHOR

Radu Tunaru is Professor of Finance at Kent Business School. He has been working in quantitative finance since 2000
and specializes in structured finance (credit risk), derivatives pricing and risk management, and in financial engineer-
ing in general. He holds a PhD in Statistical Modeling from Middlesex University in London and a PhD in Probability
and Statistics from the Centre of Mathematical Statistics of the Romanian Academy. He has published over 40 papers
and book chapter contributions and received 5 best paper awards, including the best paper award at the 2010 WHU
Campus for Finance Conference in Vallendar and the best paper published in 2011 in the European Financial
Management Journal. His latest research is in property derivatives, Bayesian models for uncertainty in finance,
numerical methods for options pricing. He serves as an associate editor on the board of Journal of Portfolio Management. His career
includes also working for Bank of Montreal and for Merrill Lynch, where he was a vice-president in structured finance EMEA RMBS. Prior
to his current position he was at Cass Business School in London.

24 INTELLIGENT RISK AC A D E M I C PA R T N E R P R O F I L E
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
PRMIA LEADERSHIP BOARD OF DIRECTORS

MATTHEW CHEN MOORAD CHOUDHRY DOMINIK DERSCH

Vice President and Head of Banking Treasurer, Corporate Banking Division, Principal Consultant,
at SunGard in China Royal Bank of Scotland Dominik Dersch Beratung
PRMIA Involvement PRMIA Involvement PRMIA Involvement
■Regional Director, Beijing Chapter ■Steering Committee, London Chapter ■ Regional Director, Munich Chapter
■ Global Council of Regional Directors
Term Expiration: 2014 Term Expiration: 2015
■ EMEA Regional Director Committee

■ 2005 PRM Focus Award Winner

■ 2007 PRM Candidate of the Year Finalist

Term Expiration: 2014

COLIN LAWRENCE ROBERT MARK OSCAR MCCARTHY

Director, Risk Specialists Division, Managing Partner & Chief Executive Strategic Risk Advisor,
Financial Services Authority (FSA) Officer, Black Diamond Risk ABN Amro Markets
Visiting Professor, Risk Management, PRMIA Involvement PRMIA Involvement
Cass Business School ■ Treasurer, Executive Committee ■ Secretary, Executive Committee
■ Chair, Finance Committee ■ Regional Director, Netherlands Chapter
PRMIA Involvement
■ Co-Author Associate PRM Textbook — ■ EMEA Regional Director Committee
■ Steering Committee, London Chapter
■ Vice Chair, Executive Committee Essentials of Risk Management ■ Former Deputy Regional Director, London

■ Former Vice-Chair, Executive Committee ■ Former Member, Education and

■ Member, Ethics and Professional
■ Former Chair and Founder PRMIA Standards Committee
Standards Committee
Blue Ribbon Panel Term Expiration: 2013
Term Expiration: 2015
Term Expiration: 2014

FARUK PATEL
Manager, Investment Risk, Alberta Investment BARRY SCHACHTER CHAE SING WONG
Management Corporation (AIMCo) Director, PRMIA Senior Vice President/Head of Asia
Business Risk Management, Marsh
PRMIA Involvement PRMIA Involvement
■ Co-Regional Director, Edmonton ■ Education Committee PRMIA Involvement
■ Former Chair, Regional Director Standards ■ Co-Chair, Publications Committee ■Former Co-Regional Director,
& Support Committee Term Expiration: 2013 Beijing Chapter
■ Former Co-Regional Director, Montreal

Term Expiration: 2015 Term Expiration: 2013

B OA R D O F D I R E C TO R S O C TO B E R 2 0 1 2 25