A Simple and Consistent Approach to Concentration Risk
Using a new data model and methodology
About
Multiple regulatory regimes across North America, Europe and Asia require financial institution monitor the concentration risk of products held on the firms’ books and record, and for clients where the firm has been involved in the transaction whether on a Principal or Agency basis.
North American regulator
“A capital charge to the firm is introduced for any concentration by the firm exceeding a specified threshold for any single security or group of related securities of the same issuer ("the security") carried by a firm in inventory, purchased or sold by clients….”
European regulator
“Firms should be especially prudent regarding credit risk: exposure of the client’s portfolio to one single issuer or to issuers part of the same group should be particularly considered. This is because, if a client’s portfolio is concentrated in products issued by one single entity (or entities of the same group)…”
Asian Regulator
“…authorized funds which may have investments of more than 10% of the fund’s NAV in securities issued and/or guaranteed by a single sovereign issuer which is below investment grade (“non-investment grade securities of a single…”
“…take into account the proportion (of the issuer) represented by the customer’s investment in a particular investment product to the customer’s total assets…”
My Role
Product Manager for the risk monitoring enhancement, my main objective was to find opportunities for reducing negative customer feedback.
I conducted research into the relevant regulations, reviewed all customer complaints logged in the past 9 months and solicited feedback across multiple functions.
In addition to translating regulatory requirements into system checks, being the author of the new methodology and data model. I had additional responsibilities as the Product Owner overseeing the end to end implementation that included the training of sales, advisory and surveillance.
The Journey
(Re)Discovery
Coming in as the Product Manager for an existing product, the hardest thing was getting up to speed with nuts and bolts of the product, the history and the current roadmap (if there was one).
Complete understanding of the product was insufficient, I needed true knowledge of the intent and ultimate vision of the product, the problem it was intended to solve, and if it achieved its purpose.
My discovery led me to 3 key themes that was common across the industry.
Inconsistent interpretation and measurement of Single Issuer
The was no common consensus on how to monitor single issuer concentration and agreed data point of what an “issuer” is. In some instances, multiple possibilities exists.
Workaround was less than ideal
The industry work around practice had been to monitor for concentration of a single security which was less than ideal.
Concentration could not be aggregated across the whole portfolio
The implementation failed in its objective of holistic assessment.
I had to relook and rework the roadmap
Multiple Personas
Being a regulatory tool, and the product being an existing one, user personas were less relevant. I used “Job Stories” in crafting the solution
There is a really good article on this found at this link: https://www.intercom.com/blog/using-job-stories-design-features-ui-ux/
Roadmap
Once I had an understanding of the journey, my next step was to combine the immovable regulatory requirements with the feedback to rework the roadmap. Here is a recreation of the strawman roadmap I used to obtain Buy-In
The Product (Solution)
A quick fix would have been to adjust the alert thresholds and to manually aggregate the risk exposures off system. However, just focusing on these 2 aspects was a unbalanced approach. We had to improve the fundamentals of the implementation.
The majority of the roadmap would have been useless if the rootcause was not resolved. For that, I authored a new methodology and data model
Data Data Data
Given the risk measure was on the probability of company insolvency, I turned to the risk ratios used in fundamental company analysis. By applying the same concepts, I arrived at a new datapoint.
This datapoint remained consistent across the various types of securities a company could potentially issue. Because public companies make this information easily available, the datapoint is easily obtainable.
With the exception of one off events, company structures remain fairly static thus removing the need for constant monitoring and updates.
The Refined Roadmap
Data testing identified opportunities for additional improvements.
Metrics
In all, end to end implementation took 9 months. The biggest challenges were:
Data verification and back testing.
Approval of the methodology. The messaging had to be tailored specifically for each stakeholder
Testing and release cycles.
Success was measured by
Error Rate: 90% reduction
Amount of client related complaints dropped by an estimated 85%.
Ongoing cost: Negligible. The data model and methodology is streamlined and was incorporated into existing processes.
Regulatory expectations: Successfully met. The monitoring methodology is more sophisticated because it aggregates across a portfolio yet is not overly conservative that business would be adversely affected
The new measures are true holistic representation of portfolio risk
Key Takeaways
Multi Discipline
Sometimes the solution can be found in a unrelated field, this was a great opportunity to use my knowledge across various disciplines such as portfolio theory, company analysis, data mining and analysis to formulate a theoretical methodology.
Given the time constraints, being a technical product manager was especially important here to have one person leading product who can not only answer the “why” and the “what”, and who can also engage in the “how.”
Meaningful testing
User A/B testing is meaningful only if the rootcasue is fixed.
Results of back testing needs to be reviewed and having a devil’s advocate is absolutely necessary.
Dig Deep
It is not enough to implement policy and regulation based on the superficial wording. Majority of the effort needs to be focused on the intent and background leading up.
Guiding through connection
Strengthening further my innate abilities to connect with people from diverse backgrounds and different ways of thinking.
How important it was to have out of office testing feedback in a comfortable or familiar environment.
Please reach out if you want to know more about the datapoint and the unifying methodology.