Commentaries and insightful analyses on the world of finance, technology and IT.

« Op Risk: Containing the loss continuum | Main | An Incentivized Anarchy »

Blurring Risk boundaries

It's been a while since my last post - I was travelling to Europe, which in recent times could turn out to be quite nightmarish. Infact, apart from leaving the airlines drooling in losses running into billions of dollars, the volcanic ash did pose a big operational risk for the project I was travelling for! - Quite a loss continuum, as I was remarking in my previous blog, but this time, just not within one entity, but across several industries like tourism, air freight and...of course, insurance (I'll get to that in a bit). Businesses which rely on airways for distribution of products, would have met with, in financial terms, an "un-modelled scenario". Interestingly, it bears quite a semblance to the credit crisis, where the CDOs and CDS' resulted in an un-factored systemic risk, owing to their risk cascading nature, which I spoke of in my initial post.

Being a supervening impossibility, the insurance companies weren't obliged to pay the claims to the airlines, but they did, in an attempt to reduce their reputational risk exposure. Now, the pertinent question is, where would these costs be booked? Clearly, the airlines couldn't sue the insurance companies, which rules out the Oprisk loss head, leaving it to something strategic, aimed at preserving reputation, client relationship and future revenue streams. Back in the Banking and Basel world, the problem at hand draws sharp parallels to the recent spark of discussions that the economic crisis was accentuated by operational risk sources. Keeping aside the Basel approach, the simple question is, "Where do we draw the line between various risks?" How to separate a poor lending choice (Operational) from a genuine default (Credit)? How to distinguish the voracity to make profits or poor investment choices (Operational) from sudden market fluctuations (Market)? Before this can be answered, we need to understand, who makes these decisions - More often than not, the person recording it is the one responsible for the loss itself - So much so for decentralisation!

The whole purpose is to embrace the "once bitten twice shy" phenomenon, particularly given the same capital charge irrespective of classification into credit, operational or market risk buckets. Furthermore, it also provides the business case for an operational improvement, which would else be sunk, much like the loss.

It would only be fair to say that risk culture, structure and process of the organisation could either make the deal (or) break it - no greys - just black or white! I would be delving into key areas under each of these in the coming days. In the meanwhile, let me know what you think...


I have been reading your past blogs, It is interesting to note such critical business observations and remarkably they are very true; while Basel did have a classification logic to demarcate certain loss types - there was no incentivisation to undertake the same, since the end effect on the capital remains the same! This clearly implies why we cant expect better risk management just by complying with regulations. It needs to be within the orgainsation to realise their significance and potential benefits.

Agreed that decentralisation is indeed beginning to work at cross-purposes; what do you propose in that context?

@ Abhinav: Incidentally, my next blog would be focussed on decentralisation! Stay tuned...

Do you suppose external loss data may help tackle the situation to an extent? I am quite divided myself on that thought!

Given that the only the loss can be recorded and not the 'nature', audits would be nothing but futile in this case. Sadly so, leaving much to the judgment of an individual. Lets see what Basel III manages to conjure up in this area

I contend that it be called an oprisk syndrome...the typical beauty of it, one problem snowballs into multiple others which it what makes it complex and intriguing and ofcourse in a utopian world makes analysis complex, which is why organisations resort to the simpler route, book it under another head and be done with!!!

I think you have a point here - We focussed more on building an AMA model that we failed to take a wholesome view of the process itself compromising on missing events or submitting to mis-recorded data in the process

@ Dani - I would say again that its systemic. More data points with each having a likely proportion of mis-recordings perpetrated by their own botched (??!!) internal processes, which the external loss data source may most likely not enlighten us on. Theoretically, if quality of controls and process effectiveness is a metric in scaling down the data (which isnt currently the case as such info isnt available), then such an activity may produce some benefits. At the current state, my take would be a NO.

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Please key in the two words you see in the box to validate your identity as an authentic user and reduce spam.

Subscribe to this blog's feed

Follow us on

Blogger Profiles

Infosys on Twitter