Ahead of the curve podcast

Navigating Complexity: FMS, Risk Management & the Power of Open Source

July 14, 2025

15 min

Overview
More episodes

In this compelling episode of Ahead of the Curve, Roland Stamm is joined by Maeve Gear from FMS-SG to explore the intricate world of winding down complex financial portfolios in the wake of the global financial crisis. Together, they reflect on their shared history at Depfa Bank, the impact of the Lehman collapse, and the regulatory shifts that followed—including the rise of central clearing and the introduction of the Uncleared Margin Rules (UMR).

The conversation dives deep into the challenges of managing long-dated, illiquid, and structured trades, and how FMS-SG leverages the Open Source Risk Engine (ORE) to meet valuation, risk, and collateral management demands. Maeve shares insights into the technical and modelling hurdles her team faces, and how ORE’s transparency, flexibility, and cost-efficiency have empowered them to streamline operations and maintain high standards.

Whether you're a risk professional, quant, or simply curious about how open-source tools are transforming financial infrastructure, this episode offers a rare behind-the-scenes look at innovation in action.

Hello and welcome to our latest episodeof Ahead of the Curve,a podcast from the Post Trade Solutions division in LSEG, which gives youan introduction to the derivatives world.My name is Roland Stamm,and with me today is Maeve Gear from FMS-SGand we're going to explainwhat that actually is, and what that means.Maeve is heading the teamfor quantitative analysis and valuation in FMS.We both go back, actually, to our common banking days.We used to work in a bank called Depfa Bank.Were responsible for pricingand representing the derivatives portfolio thereand that was before the financial crisis,a long time ago.Then Lehman happened.The Lehman default happenedand I still remember,a lot of panic and activity, from that.Right and you probably do, too.Yeah. Yeah.I remember us all working late into the night,and the traderstrying to re-hedge all those trades,trade after trade and try to terminateall those trades as well.I also remember we got free pizza that night.So,You win some, you lose some.So the result of that was that,regulators across the world decided to tighten the screwsa bit on regulation, and as a result,we got clearing, wherever possibleto do central clearing and also we got margining rules,UMR being one of them,the uncleared margin rules for initial marginand the company, after the banking experience,I worked for was a consultancythat helped building the tools to calculate marginsand that was then, turned into a serviceand the service is based on a softwarecalled Open Source Risk Engine or ORE, in shortand that's what we want to talk a little bit about.We have talked about this in previousepisodes of this podcast before,by the way,for our audience,there is a new release that just came outa couple of days ago of OREso please have a look at it.It's the 14th release since we started in 2016,and now to get into what FMS actually isit's one of those famous very long German words,but there's a good English translation.Why don't you explain to uswhat FMS-SG stands for and does?Yeah, well, you can probably dothe German word better than me,but I'm going to just say it in English.FMS stands for Financial Market Stabilisationand it's the German Government's institution,for winding up the portfolio,which was taken over from hyper real estateduring the financial crisis at that time.So the aim was to be,is to really point it upin a profitable manner to get the best possibleresults for the German taxpayer,and it's obviously very challenging because,as you will recall, it's a very complex portfolioand parts of it are long dated and illiquid.There's been a real focus on the FMS side on reducingthe risk and complexity in the portfolioand luckily or thankfully,our colleagues on the asset management sidehave been really successful with that so far.What would you say are the biggest challengesof winding down such a big and complexlong dated portfolio?Well there are many challenges.I suppose I speak more fromthe risk perspective and for me,because I'm working on the quantitative analysis side.So there are particular challengeson the structured tradesand there we have to come up withan independent market evaluation,even thoughthere isn't really a marketor liquid one anymore.These were products which were tradedbefore the financial crisis,and they're not really popular anymore at allfor good reason generally andin particular there’s been,for the hedging derivatives,which are collateralised,which obviously became so much more importantafter the financial crisis, as you just said.For those rates, we need to come up with a market valuation every single day,and we need to be able to agreethose valuations with our counterpartiesreally reasonablywell in the collateral management, process.Our counterparties arereally big multinational bankswho have lots of resourcesto put into that and we obviously don't.So that can be tough work for usand to do so in a cost efficient manner as well.That's on the one side, a kind of a technical challengebecause a lot of those kind of structured trades,we can't actually put them into our portfolio management system.either we can't capture the trades thereor sometimes we can capture them,but they don't get valued very well,we found the valuations to be unreliable.So for those trades then we have like a separateexternal pricing engine which we usean external pricing system.and then we feed the the values and the riskfigures for that back into the portfolio management system.That's the technical side of it and obviouslyon the modelling side of it, we have also their challenges,as you will recall from those days.I mean, there arestandard market models for all of those tradesfor those kind of products.So we don't have to reinvent the wheel in that sense,but the models themselvesare very sensitive to the input parametersand some of those input parametersare either unobservableor the markets data is also illiquidso that's fairly challenging,and also over the last ten yearswe've had a lot of market changes as well.We had the move it to negative interest rates,which we all thought couldn't possibly happen,and then we had the decommissioning of LIBORas well and,all of those have to be reflectedas well as infrastructure trades,so that's kept us busy over the last while.I can imagineand we went part of the waybecause while I was working for that consultancy firm,I remember thatwe had some projects together even before the one thatwe want to talk about In particular.So I mentioned ORE in the beginning,that is a full fledged risk system.It is not just used for initial margin calculations,like I mentioned,the service at Acadia and our Post Trade Solutionsis doing for our clients, but it also offers otheranalytics tools like XVA calculations.So the value adjustments of various forms for funding,cost for credit,cost for margin cost, etc..But it also supports the pricing,just the pricingof very structured, very complex trades, via its script to trades engineor scripted trade moduleand maybe you can tell usa little bit about the kinds of structuresthat you were particularly interested in?I don't know how deeply you may go into thatand how how public or un-public that is.I mean, some of it is publicand some of the simpler structures arefor instance, the LOBO trades that we have,those are essentially Bermudan option trades.Then there are also otherformula based pay offswith interest rates, inflation linked pay offs.And inflation is in particular,as you said, problematic because they are so long datedand the tend to accumulate up,right up and up and up.and that means thatthe collateral, amount also explodes kind ofand that is obviously something thatneeds to be taken care of,as you said, you have to agreeon a daily basis with your highly sophisticatedcounterparties in the market right.How could ORE help you,in handling those trades?Or why did you choose ORE in the first place,let's say.We were already familiar with the ORE, as you say.It also does XVA and we hadused it in some limited capacityalso for CVA before, but also for vanilla trades.So we had some familiarity with it and been able in recent years to actually reduce some of thethe complexity in the portfolio.So we had reduced the numberof currencies and the number of models.So we didn't have quite as many as we used to doand with that in mind, we were really looking aroundto see, if we could find a toolthat would help us maintain the quality standard that wewanted to achieve,but at the same time also reduce the operating costsfor this external pricing system that we had.The Open Risk Engine has obvious benefits.I mean, the number one is thatas an open source, it's free and there's no license costs.So you can't beat it on price,but I think the other major advantage is,of course, it's not a black box systemand you have full access to all of the codeand that means then for us, then we don't have to bereliant on someone else to explain the behaviouror why it's giving out certain kinds of results,but we can just go inand debug it ourselves and figure it outand that gives you a lot of confidence alreadyin the figures that you're producingand that justfeeds into your quality, which in itself also feeds it intocost efficiency over time.So that's I guess the background of whywe were looking at it.I think the other thing you mentioned earlierwas the scripted trades framework.When we originally startedlooking at the Open Risk Engine forcovering our portfolio,it was like mid 2023and that hadn't been released yet,but then late 2023,the script trades framework was releasedand that really gave us the possibility to coverthe whole of our portfolio and I think that was reallycrucial as well,because if you have those one or two little tradeswhich just have to be handled exceptionally,then that really just drains the resources over time.So that gave that flexibility as well.Then we said we just, we try and go for it.Would you say it was a success?Absolutely. Yeah.Really happy overall.When we went for it,we had a very, tight timeline.We really committed to itand in the end, we had to do the pre-study,the migration and the go-live all within one year.We werewe're very lucky to have an internal teamfirst of all who werereally enthusiastic about it and really committed to it.But we also we're very happyto have some support from your teamand I think thatat the start,then that special knowledgeon the scripted trades frameworkand then a few, a little tweaks here and there to thecodes and the functionalitythat really kind of made that difference for us that made it really work for our needsand particularly then on the modelling side,especially when you talk about the inflation trades as well.That expertise really came in useful for us.I'm really glad to hear thatand excited to hear that it's a successthat your reconciliations with your counterpartiesare working and calculating risk numbers and everything.So that's, really good.It's a work in progress.I mean, there's still a lot of improvement.Well, I wouldn't say a lot of improvement,but there's, of courseyou always have to keep monitoring these things andyeah, markets of course, are changing as wellbut I think the important thing is that we really,you have a flexible tooland you have something that you canthat you feel comfortable and confident in usingand I think that's what we've achieved,with the Open Risk Engine. So the scripted trade framework was originally alsodesigned for the service that I already mentionedfor the initial margin calculation.So it's not only coveringthe kinds of products that you mentionedin particular, inflation, but also all sorts of otherhybrid or highly complex tradeslike equity plus interest rates.FX plus interest rates, those kinds ofhighly structured trades, as you said,that were very popular before the financial crisisand in some areas are still very popular in Asia.You still have many of these equity and FX linkers in placeand so we see lots of these in our service as wellbut it was really great to see an adoptionin a productive pricing system internally inan organisation outside of our client basewhere we do this as a service on a daily basis right.So that's really very exciting for us to haveand we are continuing to do that.There was a press release, just recently that we dida similar thing with our sister company,Quantile, also within the Post Trade Solutions group.Where we also replaced, third party product for structured pricing and we are continuing that inside the LSEG group,but of course, it's even betterwhen even more important if our clients do thisand so thanks for sharing that,war story or whatever we want to call it.So for our viewers and listeners,maybe just a little bit of background, that we also havea lot of resources for learning how to use OREor it's not just thatwe have to come in and tell you all about it.You have a very large number of resourcesto find out for yourselveshow to use ORE, and how to make the best out of it.It is a very complex system, no doubt about it,because it does so many things.It's very flexible.So it has many, many levers and buttons that you can flickand change and switch on and off.But it is doable,as Maeve just gave testimony of,and so we have somethingcalled the ORE Academy, which is a YouTube channel.You can find some introductory videos thereto see how ORE gets installed,how to run a simple pricing and things like that.We have a large user guide, which you can also find onlineon GitHub, where you also find the ORE resources,and then of course, there's usalways there to help you if you need helpand that concludes our podcast for today.Thank you for listeningand hope to see you in the next episode.

Navigating Complexity: FMS, Risk Management & the Power of Open Source

Overview

Listen to the podcast

Also available on

More episodes

2024 Year in Review – Open Source Risk Engine (ORE)

Navigating Counterparty Credit Risk & NBFI Oversight

Market Outlook 2025 - A focus on the OTC Derivatives landscape