I would go straight to pdf on this stream of conciousness waffling piece of drivel from some years ago ... (sic) .....before I hit the whisky ....... while its printing you can enjoy this view of an RAF Tornado on excercise over Loch Sunart ... & yes this is the one that includes the geeky bit about product hierarchy modelling @ the end ... & the 'Banking Transparency 101' insert too; - enjoy!

The New Banking Transparency is inevitably DIY

Sunday, 25 October 2009

Accounting Standards are firming up (& intensifying) http://www.asymptotix.eu/content/financial-instruments-replacement-ias-39 . Basel 2.75 and CAD 3.whatever are here, now, the Liquidity Reporting Framework is on your desk. It all has to be done by Spring. You are not certain that those squadrons of postgraduate smart phone addicts are delivering anything which you understand as making even a contribution to the process of getting you there. So you chop and change project titles, realign, hire-freeze but still no matter what you do, the arrival of an operational deliverable, seems further off. The data modeling of the product table and its related entities was always known as an absolute minefield, you can't do Bank-BI without an integrated Product Table - Customer doesn't matter. Product Table Data Design is different from Customer domain modeling in one important respect; it is possible with product to have a bright analyst come into the office one morning and re-design everything you have been thinking for the last three weeks, which is less typical of Customer. Product is where the math is, where the relationships are, in a way the GL is an abstract of Product, filtered by Business Unit. That was the way UBS looked at it on the GEAR project (LINK), regarded by many as the benchmark solution architecture for bank-accounting & BI. The complexity of the product table & its integration with the GL hierarchy is what makes process transformation towards transparency difficult and lengthy. It is close to rocket-science. The big software vendors rattle off the buzz-phrase strap line; "Integrated Finance & Risk"; One software package cannot do it all for you.  Transparency is inevitably DIY. You have to know your set-theory, you have to know when you are dealing with the inclusive intersect and the exclusive.

 The accounting profession is going through a transformation, a generational thing, right now, conservatively, slowly, silently, the young turks say lumberingly; I heard an apocryphal story the other day of a postgraduate class (in the accounting department) returning this month and in one tutorial, the post credit crunch debate became heated to the point that handbags may have appeared.

 Look at Societe Generale http://www.societegenerale.com/sites/default/files/documents/CP_Q209_VA.pdf or UBS (referred elsewhere on this website http://www.asymptotix.eu/content/ubs-transparency-financial-reporting-setting-standard ) to see how transparency is done.  They both present quarterly results clearly specified as combined Basel II and IFRS compliant. For UBS and SocGen the heated argument in the postgrad-accounting workshop is over! Transparent Reporting for banks is a combination of Basel II and IFRS standards compliance and is only going to intensify further as the inclusive / exclusive slide rule moves between accounting values and econometric ones as the standard metrics of bank performance. I have said above that integrating the product table and the GL is close to rocket science for systems analysts; layer on top of that the requirement to have a full blown, industrial strength Predictive Analytic infrastructure. The 28-sheet EXCEL "tools" which put a shaking little toe in the odd monotonic extrapolation will not do in the future of banking transparency. The Command and Control centre, the Situation Room of the future for the Bank Executive Officer is going to include as many essentially predictive numbers as it does historic, just to keep that executive ahead of what the business is reporting to the market; Market Discipline is taking over as the supervisors put the enforcement pressure on the "big boys". We are moving towards the blueprint painted in very high level but presciently by Claudio Borio and Kostas Tsatsaronis of BIS in their paper of 2006; Risk in financial reporting: status, challenges and suggested directions; http://www.bis.org/publ/work213.htm

 The set theory of the Taxonomy of Banking Transparency is first of all widening to include what are essentially statistical estimates and some arithmetic interpolations of statistical estimates. For example, with a good Predictive Analytic toolset a bank can statistically estimate its RWA far faster than the arithmetic approximation of the Basel II Pillar One capital rule; that statistical estimate is also available on demand, likely to be more accurate & is certainly more useful for further analytics.

 Of course the banking transparency taxonomy includes accounting numbers but the problem is well known in relation to banking, the standards are not agreed, they are very fluid and very political. All one can say is that the clear direction of travel is towards incorporation of data analytics, calculated values and yes, predictive estimates, far more closely aligned with an econometric background than an accounting one. Certainly if you read any of the research papers being published by the top accounting schools, the next generation of accountants is going to be far more expert in statistical techniques than the current one (example: http://www.the-financedirector.com/features/feature54627/ ).

 Just as the pressure from The Treasury Select Committee in the UK is finally beginning to get through to the boards of the UK banks (since February, 7 months) to replace executive board members who have no banking expertise so the skill re-shuffle inevitably filters down through the bank executive ranks. This is an aspect of the challenge to 'the line' right now. The situation is further confused by the generational conflict already between the old guard and the new guard, the transparency wonks (whose roots are closer to the postgrads) and the "standards maintainers" summed up in this quote from a CFO taken from the paper referred above;

"Boards don't understand IFRS and as a result are becoming increasingly disengaged on financial reporting. That is a bad thing for our profession and the future quality of people it attracts. The relevance of the auditor in the boardroom is increasingly under threat."

 Transparency should not be about the relative importance of a profession, it has gone beyond that anyway, IASB and BIS have ceased feuding and the European Commission, by approving the holding of all of that toxic zombie debt can do what it likes to banks today & thus Reporting Requirements forcing banks to turn transparency cartwheels are going to continue to roll off the publication printing press. Its "all hands on deck" right now in Banking transparency, its an APB! In case you hadn't noticed there has just been an "Industrial Revolution" in banking!

 The CFO and the Accounting Profession has got to get with this Predictive thing and work with the Statisticians and the Economists on the wider set of challenges which this greater transparency sets. The bigger universe of reporting data should be regarded as the centre of the command and control of the bank, external forces of market discipline pouring in through the open supervisory window, to which the crisis exposed the industry; demand of every responsible executive now, that they are aware of a headline subset of information which is about 50 / 50 predictive / historic. Possibly 60 / 40  econometric / accounting?

 In the 1^st snorts of reaction to the consensus I think it was of the UK Treasury Select Committee to require bank executive directors to have some form of banking experience or relevant professional qualification, one point was made which made sense & that was 'where are we going to get all these people who are experts in fields relevant to banking?' I think I have made the point (above) that we need people with the expertise of qualified accountants up to date with the latest standard PLUS people who understand Predictive / Statistical Techniques to understand what the numbers mean. Rocket Scientists. A Senior UK banking executive told me at our last meeting this year; 'Engage with the Line, John, Engage with the Line!' I didn't understand it at the time but I do now; another bank in the British Isles which I know well, recently proudly boasted of having a team of PhDs in the Risk Department, that was the bank which in 2005 looked with white faced horror at some math on a power-point. Banks need ACCOUNTING+, Integrated Finance and Risk Reporting as SAP invented the konzept. In the line, how many experts are there, who really understand all of the combined transparency schedules for banking today? It's a tall order to say that you do. Thus to design a solution in house by climbing that sheer IP cliff i.e. waiting until your organization looks like MIT before starting to integrate something which might address the reporting requirements is a non-starter, in my view, you just don't have the time. Noone does in European banking It might be 10 years and whole new generation before we have a planet-brain workforce which understands the whole gamut of banking transparency. That's if our Universities are doing anything to help us catch up with this knowledge-gap, which some of them are, example: http://www.macs.hw.ac.uk/

I know what I am effectively advocating here. I understand how powerless 'the Line' in Bank Management must genuinely feel right now, the worst nightmare has actually occurred, worse than the mathematicians or the economists taking over the bank (which is beginning to happen) now Line Management have to implement software applications which produce numbers for new Rocket Science like reporting requirements which only specialized accountants, economists or statisticians might understand, right? So Line Management in Banking right now has to take advice from geeky experts talking gobbledygook about statistical significance, slide rule techniques for impaired assets or waterfalls. Line management has to implement new systems costing several millions in record time to meet reporting deadlines. Line Management has to take it on trust from these 'experts' who could well be accused of being so obsessed with their field that only the Murciélago will do, not only what the solution needs to be but what the requirement is, in the first place. Inevitably with this raising of the bar of reporting requirements some fairly intelligent members of the team are going to be reduced to mere plumbers. But, the more you really think about this requirement the more you realize that in data terms but also in terms of those processing hungry predictive algorithms this really is a requirement which requires sports car level technology, even the plumbing is not a simple job.

 What's my point, None really! I don't have a conclusion on all of this. This is My contribution to the discussion which I know is happening in every bank boardroom in Europe. I wouldn't presume to second guess the management decision, I simply thought I would lay out the issues and air some of my thoughts. This is indeed a massive generational challenge.

John A Morrison Profile / email John