Posts categorized "Events"

Counterparty Event

I went along to a morning panel on counterparty data management on Tuesday, sponsored by GoldenSource, Avox and Interactive Data, and hosted by Virginie O'Shea of the A-Team. Counterparty data obviously has a very high profile currently in light of recent events, however the advice from the panel fundamentally seemed to be get the basics of data management right (ownership, control, consistency, quality, transparency), rather than anything radically new.

There was some debate about the possible extension of BIC (Bank Identifier Code) to be used more generally as a standard for a unique business entity identifier - this seemed to be received well but there were concerns that such an initiative would not solve the problem but rather become an addition to the already complex entity-mapping process.

The "Data Utility" from the ECB was also debated, and it was refreshing to here some negative (realistic?) things said about it, such as the concern raised by Interactive that this might involve huge public spend without necessarily understanding why a new government sponsored entity would be able to do better than existing data providers. Obviously a data provider would say that, but I have to agree, it seems there is too much focus on having a data utility and not looking at the different options for solving industry data issues (one option obviously being a data utility, but lets not pre-package the problem with a solution but more of that in later posts...).

For more detail on the event, then take a look at Virginie's blog post.

Posted by Brian Sentance | 21 May 2010 | 10:56 am

RiskMinds - from Blame to Bubble Indices...

I am attending the RiskMinds Conference in Geneva this week. Given what has happened over the past year, its somewhat intellectual name seems less appropriate than it once did, but I guess not many of us are smelling of roses on that point...

Seems to be very good attendance with the main hall full to overflowing for the first full day of the conference - unsurprisingly I think many people are looking for answers (from "what did I do wrong?" to "who can I blame?"). From a quick survey of the attendees, there seems to be no doubt that regulators and the credit rating agencies seem to be the favoured candidates to blame.

Robert Shiller (author of Irrational Exuberance) gave the openning talk on the current credit crisis and what to do about it. He made the point that behavioural finance (stock market pyschology) is becoming much more integrated with financial markets theory, and put forward the positive point that financial theory needs to be expanded to encompass what we have experienced over the past year, not that all financial theory should be thrown away (a jibe at Taleb on this point?)

Much of Professor Shiller's talk was spent on illustrating various "bubbles" in real asset prices in various markets against long run trends, usually involving a comparison with the data of the Great Depression of the 1930's, and an occaisional mention of his book (I haven't read it (yet) but I would guess it spends a lot of time on bubbles too). He is very keen on the democratisation of finance, more particularly of financial advice (it would seem that the FSA has been listening in the UK, with the recent action against commission-based financial advisors).

He also proposes greater usage of macro economic indices and related derivatives to make risks of house price falls, inflation, economic growth, employment etc more transparent to all and to allow easier hedging of these risks. He raised some eye-brows of many banking staff by proposing mortgages whose payments went down when these factors moved against a house owner (with the originator hedging these risks using futures on the indices he proposes). He was not so clear what should happen when these factors went in favour of the house owner!

One thought struck me though the talk, is that if it is relatively easy to illustrate/calculate these real asset price bubbles illustrated by Professor Shiller, then why not go further than just having indices on direct macro-economic variables and have indices based on these "bubble" calculations? If everyone could see that the "bubble" index for a particular risk factor was high then you could hedge your "irrational exuberance" or at the very least there would be a transparent indicator that a market was moving into dangerous price territory. Stupid idea? Maybe, but if it has legs please remember you heard the nickname"Aero" for the cocoa index here first!...

Posted by Brian Sentance | 9 December 2008 | 10:42 pm

Here today. Where tomorrow?

These may be the words on the lips of many bankers today, as they survey the continuing turmoil in global financial markets. In fact, this was the incredibly apposite tagline on a recent magazine advertisement for a major bank which (maybe unsurprisingly) was subsequently nationalised.

In the fluid (many would say “bloody”) landscape of financial services, with the next merger or acquisition just around the corner, it means that now, more than ever, data integration is a growing challenge. Accompanying this activity is the ever-growing need for consistency, accuracy, transparency and control of both the data itself and the movement of that data.

Data architecture itself is an evolving discipline and one approach worth looking at is data federation – deftly described in an article by Dain Hansen. Basically, the approach is to leave the data where it is but aggregate it into a single view, available as a service to your applications. It is an approach that Xenomorph has advocated for many years, going back to our founding days in the mid-90s, with the normalized database driver approach implemented in our Connectivity Services.

Hansen’s article explains both the advantages (simplicity, no need to copy or synchronize) and the disadvantages (performance) of this approach, and argues for a solution that incorporates both federation and consolidation of data. He shows that it is possible to architect a system that will provide consistency and control as well as agility.

It’s difficult to say whether better data management would have assisted the world’s banks in avoiding their current troubles, but greater transparency of where exactly their exposures lay would certainly have helped.

Posted by David Winson | 1 October 2008 | 6:28 pm

Industry Trends - Larry Tabb at Sifma NY

Main points from a good talk given by Larry Tabb of the Tabb Group last week in New York:

• The total loss so far from the sub-prime/credit crisis world-wide is around the USD 280 Billion mark – although the final figure could be as high as USD 400 Billion as in some cases the dust has not fully settled.
• With a few notable exceptions, the majority of these losses have been incurred by the major US banks rather than their European counterparts.
• The effect of the situation has hit a number of asset classes, the worst by far is fixed income where trading is down by 80% at the moment but it’s coming back. Equities – particularly US – has also been hit pretty hard but again it’s showing resilience and starting to improve
• In terms of what is going to happen, IT spend is going to decrease from previous levels for a year or to by 10% or so, but is expected to rise after 2010.
• The only areas where IT spend is expected to rise are communications and risk management.
• Communications because there is much more activity in emerging markets, particularly Asia (excluding Japan - mainly India, China and HK) and also Latin America. Connectivity is needed to the exchanges within these markets so there will be IT spend to make this happen. Also, there is going to be more algorithmic trading as people try to get more from ever-tighter markets and reduce costs through lower trader headcount and paying less in broker commissions.
• As manual trading decreases due to algo increases, brokerages may increase their commission rates to compensate and some of them are also going to introduce caps so that clients have to put so much trade traffic through them or else they won’t be interested - thus reducing numbers of clients and streamlining the process. This will force smaller firms to use smaller brokerages, thus splintering the market more.
• Risk management spend is going to increase because firms want to understand why their models failed w.r.t sub-prime situation and why they didn’t see these huge losses coming earlier. They don’t want it to happen again so they want tighter and better tools as a result.
• The breakdown by asset class in algo trading is expected to change. At the moment it is largely Equity, but the biggest increases are expected to be derivatives, futures, fixed income (particularly credit), FX and a lot more OTC to make this automated rather than manual to get these exposures off the balance sheets. This obviously means that the exchanges and banks/funds etc. need the software and equipment necessary to do algo trading in these spaces which currently are quite embryonic so there will be a lot of development in this area. Of course, this means algo engines and risk management/portfolio software will need to be much more adept at handling mixed assets and not just equities.
• All firms, banks and hedge funds, are going to place a lot more of their investments overseas, due to higher expected return and less risk, rather than the US, so all of their systems will need global capability in terms of data acquisition and trading activities, with consolidation and risk management
• Larry also mentioned that he had been to TradeTech and also one in Asia, both of which he had been involved in panel discussions with all of the Exchange Heads (e.g. LSE) – and when asked what the biggest headaches were they all said that clearing was a massive problem with the process being splintered and disparate and if the trading levels are going to increase this process must be a lot more streamlined.

Last point is very topical in news at the moment given LSE assessing whether to get involved in doing its own clearing, plus also the regulators desire to get some form of clearing in place for major classes of OTC derivatives. 

Posted by Brian Sentance | 19 June 2008 | 10:25 am

TradeTech Paris 2008 - Chi-X leads the waiting game

We exhibited at the TradeTech Paris event last week - this is a mainly equities/trade execution event and as such most of the speakers were playing the waiting game and not saying much. Everyone seems to want to see more of the new trading venues come on line before they put out any opinions of substance.

The main (only?) news item was the rapid uptake of Chi-X and its share in trading volumes against the exchanges. There was some "competitive" banter between Peter Randall of Chi-X and Eli Lederman of Project Turquoise (Chi-X saying that Chi-X is a live business and not just a "project"). LSE was there too, not doing a great job of defending the record of exchanges - too much emphasis on defending their existance based on the past rather than the future in my opinion.

Posted by Brian Sentance | 28 April 2008 | 4:01 pm

Higher quality data from the front office?

Sungard had a good event on Thursday night, with four risk managers taking the stage for a "thought leadership" seminar entitled "Regulatory Impact of Market Events" (if the advert is still around on their site, see http://www.sungard.com/ADAPTIV/default.aspx?id=4678&formAction=takeit&formid=48)

The Dresdner risk manager (Ted Macdonald, good speaker) was emphasising that data quality is a real issue for risk management, and that all participants thought that risk managers should spend more time on risk and less on validating/cleaning data (no great surprises there then but interesting to hear it validated again as an issue).

He suggested that more pressure should be put on the front office to get data right first time (as opposed to leaving everyone else to sort out the mess!), even going so far as to suggest that charging the front office for each wrongly-booked trade in the trading and risk management systems - not sure how that would go down with the trading desks, but sounds a good approach if you could agree (and unambiguously measure) these mistakes!

Seems like transfer-costing is becoming a re-occurring theme - also recently mentioned by a grid computing specialist from Credit Suisse about "metering" each desk for the amount of compute power used...anyone retraining as a management accountant out there? - sounds like the banks will be hiring soon!

Posted by Brian Sentance | 17 March 2008 | 3:08 pm