This is my fifth installment in a series in which I more systematically outline and discuss risk and crisis management processes, and best practices for building them into a business? basic organizational infrastructure (see Business Strategy and Operations ? 2, postings 333 and scattered following for Parts 1-4.)
If you were to ask most any risk management professional what their most important tools are for determining risk level, historical trends analysis and benchmarking based upon that would be among the first cited. My goal in Part 4 of this series was to at least relatively systematically question that near-automatic assumption of value, and not simply in the abstract but in terms of a very well-known real world example: the over-reliance on historical data when evaluating risk in complex bundled investments as that led to and significantly contributed to causing the Great Recession.
I focused on novelty and change as a primary reason why history was not a valid source of performance or risk prediction for these investment firms as they aggregated and assembled ever more complex and opaque bundled derivatives and derivatives of derivatives. And in the course of that I posted a conclusion bullet point that I reiterate here:
? And the biggest single source of error that they brought to the table was attempted reliance on historical data where novelty and change meant there wasn?t actually any real history to turn to ? and what business history there was did not actually apply.
And that brought me to a fundamentally important question that I finished Part 4 with as a concluding point for thought:
? If a business cannot rely on its own historical data or on industry-wide historical data for determining and setting a proper scale for risk and opportunity, what should it use and how should it do so?
Historical data and the trends-based risk assessment models that this flow of information provides, forms the basis for insurance rates and policies and for so much more. There are and will continue to be, many contexts and circumstances where historical data and trends analysis based upon it would offer real and even primary value. But when this accumulated cornucopia of data and insight cannot be relied upon for the novelty of what is to be evaluated and risk-managed, what should a business owner or leader or their risk management professionals do instead?
My goal in this posting is not to try and provide anything like a comprehensive, definitive answer to that question, but rather to outline some of the key parameters that would have to be taken into account in formulating any effective response to that challenge. And I begin with a point that should be obvious but that strikes to the heart of how and why so many major financial institutions fell into massive risk that they did not recognize as such ? until their investment systems failed.
? Opacity is the enemy of due diligence and of risk management.
These toxic investment options were assembled as if pouring unexamined and unknown quantities into a black box. And that I add included at least some very significant separate components such as high-risk home mortgages where the home buyer did not meet standard, vetted criteria for being eligible for a mortgage, that should have been identifiable as very risky from historical data.
? If you do not really understand and acknowledge the risk of the pieces of a complex product or system that you are assembling, you have no way to evaluate the overall risk levels faced from that assembly as a whole.
? And you have to analyze the component parts where possible for their risk. Simply hiding high risk elements in a more complex system does not and cannot cause their separate and individual risk contributions to simply disappear.
And this brings me to the challenge to simple historical trends analysis that I touched upon in Part 4:
? You need to know and understand the significance of new and disruptively new and when you are in fact dealing with novelty for which you do not have a historical basis for comparison. This has to enter into the identification and characterization as to risk of all key systems components.
? And this means that you need to understand the consequence of novelty and of blue ocean potential in strategy, and in its execution as well as in specific products or services offered, for how this limits the ability to risk assess from historical data.
? Analyze by historical trends where you can realistically do that but first and foremost know where you cannot do that and where other risk assessment approaches would be needed.
As a final point for this installment, even to the extent that the historically analyzable components to those complex derivatives were identified and evaluated for their risk, the scale and complexity of the derivatives they were bundled into were such that there was no baseline data to risk evaluate them from. Historical data did not offer real insight as to how lower and higher risk components did or did not balance off in those complex derivatives, and derivatives of derivatives. So they were new and without precedent and that is where these systems of trades fell apart for over-reliance on irrelevant historical data.
So in Part 4 I raised the issue of how historical data and trends analysis based on it cannot always offer risk assessment value, and in this I discussed the need to identify the specific areas where this type of standard analysis would break down, and where it would continue to work for you. In my next series installment I am going to at least begin a discussion as to what to do in evaluating systems components that do not fit historical trends due to their newness and novelty. Meanwhile, you find this and related postings at Business Strategy and Operations ? 2 (and also see Business Strategy and Operations.)
Like this:
Be the first to like this.
Sasha McHale Boy Meets World chicago bears elizabeth taylor casey anthony chris brown cam newton
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.