Black Holes vs. Black-Scholes:

I have specifically carved out a chapter for valuation, as it is never ceases to be an important part of any system, in any trading and risk management strategy.

The dichotomy of this area is that we have many basic valuation tools, including some old-school tools that became the norm out of there was nothing better off-the-shelf to use.

Even with greatly known limitations, it was the best we had. Now, in the 21st century, we are seeing more and more people moving to big data simulation – stochastic techniques that unleash the tail-risk, provide much better insights, and generally provide much better information than ever before.

We also have many companies who are building home-grown systems. Oftentimes, though, these systems take years to develop in-house, even if you buy them off-the-shelf, as someone has had to put years into development the process. The challenge for many companies is the complete lack of these tools, but also where the pendulum has swung so far that 25 risk staff are taking models to a point of diminishing returns. We then end up knowing more and more about less and less. That may be humorous the first time you read it, and the second. After building and consulting for years, this is no laughing matter.

Senior management is often confused about many ideas these models put forth, internally developed, or externally purchased. The complexity, or rather that executives do not use these models every day, makes communicating the results more challenging. Communicating what those models do, much less getting into the weeds of every last input and error term, is not effective in communicating the picture of risk.

Executives don’t have time for the details, the tactical challenges you face, or the debates the risk folks are having amongst themselves. Focus on building what I call a “Picture of Risk,” and communicate that picture. State the key outcomes with a broad brush, a few implications of two or three scenarios, and leave them with the information to digest later.

Purposefully communicating the risk and opportunity is key, not solving a mathematical problem only university professors care to discuss.

Finding the best middle-ground solution therefore, means it is better to buy than build. And it is time that many may want to stop using the old-school valuation tools and move into the 21st century using what is commonly known as auto-regressive modeling, physical optimization, and/or stochastic tools. Along with more advanced tools, crunching big-data into simulation analysis provides much better results to maximize your risk-reward profile.

Purchasing an out-of-the box system is exponentially much cheaper than trying to build one yourself. Sure, you can also request a small change to their models to fit your internal modeling approach.
This is not to say these more sophisticated tools are impeccable.

Impeccable is incongruent with human nature and the increasing change we face in all directions. For the cost of the system and implementation, these systems work very well. Getting the cash-flow-at-risk is an important attribute, for example, not just the valuation. The gross-margin-at-risk is another. Better credit results too. I could go on.

What derails some of these projects, just like C/ETRM ones, is that often the data provided is suspect and the client doesn’t know it. I have been on projects where the client believed their data was good, only to discover it wasn’t. The data has never been vetted or validated, and when the results are put into these off-the-shelf systems, ETRM transaction processing or into the big data physical optimization systems, there are plenty of cases where the data is the culprit. Glaringly, I may add.

I do not disagree that at times there are promises that are not met by a vendor, perceived or contracted. However, to find an optimal state, many will have to overcome their tendency to model and debate every nook and cranny of the models.

In summary, using 21st-century developed valuation tools is going to produce much better results for not only trading, risk, including accounting, it will pay for itself in communicating with the back-office to the boardroom.

For questions, contact Steven Berley at (303) 304-9720 or steven@frontmidback.com