🔍
The Genesis of Basel IV: Fundamental Review of the Trading Book | Numerix Video Blog - YouTube
Channel: numerixanalytics
[6]
Jim Jockle (Host): Hi welcome to Numerix Video
Blog, your expert source on Derivative trends
[9]
and topics.
[10]
I’m your host Jim Jockle.
[12]
As institutions continue to embark on implementing
the necessary changes related to Basel III,
[17]
the distant hum of what many market participants
anticipate to be the genesis of ‘Basel IV’
[22]
is looming in the background.
[24]
From a market risk perspective, all eyes are
on the draft proposal of the Fundamental Review
[28]
of the Trading Book.
[30]
Ultimately, this revised framework of market
risk guidelines could bring significant change
[35]
to both Internal and Standardized models.
[39]
Joining me today to discuss this is Satyam
Kancharla, Chief Strategy Officer and SVP
[44]
of the Client Solutions Group here at Numerix.
[46]
Satyam, thanks for joining.
[47]
Satyam Kancharla (Guest): Hi Jim.
[48]
Jockle: So, quick question, to start with.
[50]
You provided us a summary of what we believe
market participants should be thinking about
[54]
in the coming months.
[55]
So, why don’t you take us a little bit through
a background of the timeline of some of the
[60]
issues with FRTB.
[61]
Kancharla: That’s exactly the question that
a lot of people are now talking about.
[66]
What’s the timeline, when is this going
to be finalized, when is this applicable to
[73]
different types of institutions.
[75]
With respect to how it all started, back in
2009, we had what we called Basel 2.5 and
[83]
at that time it was a reaction to the global
financial crisis, it was an immediate set
[88]
of things the Basel committee felt that it
should do in order to put some additional
[93]
risk controls and additional capital charges
in place.
[97]
At that point, they also parked certain issues
for fundamental review, hence, Fundamental
[102]
Review of the Trading Book that would be done
later and since then, they’ve been doing
[108]
several versions of consolidated documents
to describe what is now being called the Fundamental
[115]
Review of the Trading Book, also being called
as Basel IV.
[120]
And there is a general, the philosophical
direction is moving towards looking at the
[128]
entire book from a positions perspective,
as well as from a risk factors perspective,
[134]
to look at what risk factors are at play.
[137]
Both for the standardized approach, as well
as the internal models approach.
[140]
Jockle: So there’s really 3 key areas that
are new as it relates to FRTB, number one
[148]
expected shortfall vs. VaR, liquidity horizons
and some of the methodologies.
[154]
Can you give us a little bit more depth on
those issues?
[157]
Kancharla: Absolutely.
[159]
One of the changes that we are expecting to
see is a move to expected shortfall as a measure
[167]
of Risk and as a measure that then builds
into the capital, as opposed to VaR.
[173]
Now expected shortfall obviously being a tail
measure, seen as something that can more accurately
[180]
capture the tail risk that exists in a book.
[184]
The other factor that we see is a liquidity
calculation that’s far more granular than
[190]
what we saw before, whereas in the past we
had a blanket number, which was a multiplier
[198]
used by Basel.
[199]
Now we have something that’s more tailored
and more granular, based on the specific liquidity
[206]
profile that we expect to see or that we see
in the market for different types of instruments.
[211]
And then there’s this whole notion of looking
at internal models and standardized models,
[217]
not at an entity level, but also looking at
it at a desk level.
[221]
So every desk is being examined, or is going
to be examined, from a PnL perspective, from
[227]
a back testing perspective; so the regulator
can essentially switch between the standardized
[233]
approach and the internal models approach
at a desk level, as opposed to the entity
[236]
level.
[238]
And then, also different risk factors are
being looked at and there is this notion of
[245]
a non-modelable risk factor, and for that
they’re going to apply a different type
[250]
of approach.
[251]
But there are certain conditions under which
a risk factor is supposed to be modelable.
[260]
And then finally, there is the move towards
incremental default risk, incremental risk
[263]
charge, and essentially incremental default
risk focuses only on default versus both default
[273]
and migration risk.
[275]
And the reason is, the Basel committee felt
that migration risk and spread risk had a
[284]
lot of overlap and they didn’t want to have
double counting off that specific measure
[290]
of risk, so there is a focus to default risk
only within IDR, incremental default risk.
[297]
And finally, there are a number of changes
on the standardized approach and there’s
[301]
also a sensitivity based approach, which is
intermediate between a standardized approach
[305]
and internal models approach that’s being
launched.
[307]
Jockle: So, we’ve had conversations around
BCBS 239 with the looming deadlines, and some
[314]
of the technology challenges, in meeting that
principle requirement.
[319]
What kind of technology issues arise with
FRTB?
[323]
And I’m assuming there’s probably 2 buckets,
one being in the realm of quantitative financial
[327]
innovations, as well as core technologies.
[331]
Kancharla: Absolutely.
[333]
And, on the methodology side, as we discussed,
we’re moving towards expected shortfall,
[339]
moving towards liquidity horizons.
[343]
With expected shortfall, being a tail measure,
in principle there’s a lot of agreement
[350]
about expected shortfall being a better metric
of risk in the book.
[357]
However, there are a number of open issues
with respect to how this tail risk has to
[362]
be measured and how it can be measured reliably.
[365]
Given that the tail we’re talking about,
literally a few scenarios in a typical 250
[371]
a day VaR scenario, you’re talking about
a small number, a handful of scenarios really
[376]
determining what the expected shortfall is
and therefore, whether we can reliably estimate
[382]
expected shortfall in this manner, or do we
really need to switch to a Monte Carlo based
[388]
approach or longer time horizon is something
that is being discussed at the moment.
[394]
Also, liquidity horizons¬–how we calibrate
them, how we measure them, how we back test
[399]
them, these are all important questions that
have to be handled.
[404]
Also, on the standardized approach, what we
are going to see is standardized approach
[410]
is going to be mandated for all institutions
and it is also likely going to be used to
[417]
determine a floor or a surcharge.
[420]
Which means that institutions have to not
only compute internal models, but also standardized.
[425]
So both have to be computed.
[427]
And, that again brings a lot of data issues
and technological issues that we’re also
[434]
seeing in BCBS 239.
[436]
Jockle: From your opinion, it was clearly
advantageous in the past, to utilize a non-standard
[442]
approach specifically as it related to capital
relief.
[447]
With this change in terms of having to provide
both the standard model and the IMM, are banks
[453]
going to see the same benefit through an advance
model?
[456]
Kancharla: I think that’s a great question,
only time will tell.
[460]
And there is a quantitative impact study that
people are expecting to see.
[464]
In many cases are being conducted internally
within institutions and that will determine
[470]
how the relative metrics will compare.
[474]
But in general, the idea is that the standardized
approach will be defined as a common metric
[483]
that will be derived from any bank, and in
some sense it provides regulators with what
[490]
the capital is going to be on a standardized
basis across the industry and therefore there
[497]
is an attempt to bring these two, the standard
metric and the internal models together, so
[502]
that there isn’t a lot of opportunity for
arbitrage, or switching back and forth.
[508]
So my expectation is that we will see these
measures converge more than we have seen in
[514]
the past.
[515]
Jockle: Well, Satyam thank you so much for
joining us and shedding light onto this topic.
[518]
And of course we look forward to discussing
with in the future and we want to hear your
[523]
feedback going forward.
[524]
It’s our goal to examine the topics that
you want to talk about.
[527]
So keep the conversation going on LinkedIn,
and on Twitter @nxanalytics.
[531]
Check back soon for the latest industry news,
blogs, videos, research and technical papers.
[535]
Thanks for joining.
Most Recent Videos:
You can go back to the homepage right here: Homepage