ISDA SIMM and FRTB CVA | Numerix Video Blog

ISDA SIMM and FRTB CVA | Numerix Video Blog


Jim Jockle (Host): Hi Welcome to Numerix Video
Blog. I’m your host Jim Jockle. As we’re quickly closing the books on 2015, 2016 promises
more costs, and that cost is directly impacted by two elements in the market. First, is the
introduction of certain deadlines as it relates to FRTB under Basel III rules, and the second
is the introduction of ISDA SIMM Methodology, which is margin for non-cleared derivatives.
Interestingly, just before Thanksgiving here in the United States, Basel just put out its
interim impact study, looking at some of the effects on FRTB. And, the report had concluded
that we’ll definitely be seeing something of a minimum 4.7 percent in capital over the
additional Basel III capital rules that have been rolled out into the banking book.
Joining me today to discuss some of those regulatory changes that have been enacted
over the past few months is Dr. Serguei Issakov, SVP of Quantitative Research and Development
at Numerix. Serguei welcome. Serguei Issakov (Guest): Thank you Jim. Jockle: Perhaps you can give us a bit of an
update of some of the changes that were enacted and proposed into the market during the summer
months. Issakov: During the summer, the Basel Committee
issued a new regulation, which is a substantial upgrade of the existing CVA capital regulation.
And, that regulation it’s an update. It’s aligned with the FRTB approach as you said,
so Basel proposed one approach which is a generalization of the standardized Basel III
capital calculation for the bank that currently used this approach. And the change was to
account for market fluctuation and exposures. The original Basel III CVA capital calculation
counted only on fluctuation in credit spreads for counterparty when computing CVA VaR, which
is the quantity on which the CVA capital is based.
Now they are basically under pressure from banks. Banks normally do account also for
fluctuations in exposures. They added this requirement to account for those fluctuations
in calculation of the capital, CVA capital. That’s one part of this proposal. And, also
they issued their so called FRTB CVA Framework, where they aligned the calculation of CVA
capital with FRTB sensitivities, and FRTB shocks, define the FRTB approach to market
data. So now it’s interesting to note that FRTB
framework serves as a unifying framework for different approaches to the calculation of
regulatory capital. Jockle: So Serguei, let me ask you a quantitative
question. I don’t know if I’ll understand the answer, but… One of the things that
we’re continuing to see different dislocations in the market. We’re seeing changes in credit
spreads in terms of synthetic credit, widening out over cash credit, we’re seeing more
and more volatility, and things of that nature. A lot of the regulation as it was developed,
especially in the terms of quantitative finance is assuming stable and predictable markets,
yet right now when you start seeing different volatility swings and things of that nature.
How does that impact some of these calculations? Assuming all things were equal in terms of
credit spreads, but yet you see a market dislocation. How is that going to disrupt some of these
models? Issakov: Well, definitely the models will
have to be adopted to be able to handle large volatility changes, volatility spikes and
different volatility regimes. But, that’s what quants have been working on all the time,
when such regimes do occur in the market. Definitely, increasing credit spreads would
increase the capital charges. But, that’s basically how we can just react and account
for those fluctuations in the market. Jockle: OK. Very good. Thank you. Just turning
over to some of the SIMM Methodology. What are some of the challenges that people are
thinking about as they are looking to implement this?
Issakov: Well, the SIMM Methodology which was issued by ISDA, it’s a Standardized
Initial Margin Model. It looks like it is going to be accepted in all regions of the
world. And, the Methodology is pretty simple in the sense that if you are able to compute
sensitivities according to the FRTB approach, then, the rest is pretty straight forward.
But the correct computation of those sensitivities is a real challenge. You have to be able to
compute them in an accurate way. And this is where models, in fact front office models,
become more important for the computation of those sensitivities. Jockle: And obviously this this will ultimately
lead into the introduction of the MVA, the margin Valuation Adjustment. Perhaps, what
should people be thinking about as they are incorporating the concepts of MVA into their
total workflow? Issakov: Absolutely. MVA, is similar to KVA,
which is the cost of your capital throughout the life of your portfolio of trades. For
margin, when margin requirements are applied, you have to be able to compute your margin
requirements again throughout the life of your portfolio. And, well, based on the approaches
for margin calculation itself you can simulate the calculation, you can simulate the MVA.
Well, MVA is the cost of margin. The Methodology is similar to the calculation of KVA. In fact,
you can say that even for MVA based on initial margin for bi-lateral trading, mathematically
it’s very similar to KVA for market risk capital. Mathematically, it’s almost identical,
in fact. So, if you know how to compute one, you know how to compute the other.
Jockle: Well, finally something easy in the world of quantitative finance.
Issakov: Yes. Jockle: Well, Serguei I want to thank you
for joining us. Of course, stay on top of all the things that Numerix is putting into
the marketplace. Look for our new paper on FRTB that will be available on numerix.com.
Thank you so much for joining. And of course on the Numerix Video Blog, it’s our goal to
examine the topics that you want to talk about, so please keep the conversation going on LinkedIn
or on Twitter @nxanalytics. Thank you so much for your insights on the topic.
Issakov: Thank you Jim. Jockle: And, we’ll see you next time.

Leave a Reply

Your email address will not be published. Required fields are marked *