Financial Services: Predictive Analytics with Blockchain and AI
The changing consumer experience at an exponential scale is one of the contributing factors creating risk and compliance disruption in financial services. The “merging” of banks and non-banking entrants (including the “BigTech” e-commerce giants) through marketplace platforms and new business models will introduce regulatory changes from data privacy to the core credit, market, and operational risk pillars of “systemic” risk exposure and classification for the industry. The size of the bank alone in terms of assets is no longer sufficient to identify systemically important banks or non-banking entrants moving into the industry with a banking license. Risk and compliance will be disrupted in this context, as a multifactor approach will need to further evolve when considering and determining a new systemic importance baseline beyond just asset holdings.
Regulations typically use bank assets to identify and categorize such firms, but asset size alone does not equate to risk to financial stability. An alternative approach for identifying global systemically important banks (G-SIBs) relies on a multifactor approach and multiple measures, not just size. Thresholds using multiple metrics can give a more nuanced view of a bank’s systemic importance than asset-size thresholds. Global regulators that are members of the Basel Committee on Banking Supervision agreed to develop a methodology with the following criteria for identifying G-SIBs:
- Cross-jurisdictional activity.
Twelve systemic importance indicators determine the scores in the five categories. Size is measured by total exposures, not assets.
The Barclays team of Michael Cohen, Dane Davis, Nicholas Potter, and Warren Russell “define a commodity ‘black swan‘ as an extreme event or dynamic that market participants, including ourselves, are not currently pricing in.”
When looking at exposures, it is also important to look back at historical black swan events in terms of credit, market, and operational risk impacts to financial institutions. The effects can be compounded in “perfect storms” of events that can also be engineered by bad actors/humans in terms of large scale fraud for financial gain to manipulate market outcomes. Money Monster starring George Clooney tells the story of a chain-reaction set of events that cover market, credit, and operational risk, with market manipulation and outcomes created by macro political “jiggering” and election-fixing.
Could an AI-engineered solution combined with Blockchain and Smart Contracts prevent this type of scenario and other variations of this scenario?
The required assessment to determine systemic risk exposure will require advancements in financial innovation for the global financial system such as Blockchain with smart contracts alongside the convergence of AI and predictive analytics to determine exposures before they happen. The sheer interconnectedness brought about by new business models and merging of industries will require a better mousetrap for identifying much larger complex systemic risk across these digital financial marketplaces.
The proposed solution here is to develop a 3-dimensional or 3-pillar predictive analytics risk framework and model that to start with is Basel-based. It would cover:
- Credit Risk
- Market Risk
- Operational Risk
The model will leverage data from black swan events such as key results and outcomes from the 2008 Financial Crisis and mortgage industry crashes; hurricanes such as Floyd, Katrina, and Harvey in the US; the 2011 Japan tsunami; and other major natural disasters.
The model would be constructed on a Blockchain using smart contracts, which represent normal transactional and network data and show the impact on credit, market, and operational risk dimensions. It could potentially be developed and built to represent impact for large public institutions (public chain) and smaller institutions (private enterprise, “behind the data center firewall” sidechains). Ideally, existing archetype network topologies based on each institution type, along with transaction types could be used to set up the Blockchain structure and nodes. Then by analyzing network data and traffic for black swan events compared to normal traffic and processing (as represented in the smart contracts), outliers could be identified. For example, is there an overload of abnormal transactions outside the smart contract parameters that caused error conditions and faults?
This information could be found and used by AI to update contracts on the Blockchain and predict with better certainty the impact of black swan events.
Who Would Benefit?
- Regulators/Regulatory Bodies
- Central Banks
- Governmental Financial System Research organizations
- G-SIBs (Globally Systematically Important Banks)
- Hedge Funds
- Credit Unions
- GAFA (Google, Apple, Facebook, Amazon, TenCent, Ali)
Below is a representational overview of SupTech (Supervisory Technology) that illustrates the progress made by various country- and regional-based governmental supervisory agencies in comprehensively monitoring and managing risk at macro and micro levels.
Source: Innovative technology in financial supervision (suptech) – the experience of early users by Dirk Broeders and Jermy Prenio, July 2018
Monetization: Developed SW/HW and appliance stack, and offered through platform as a service. Ideally hosted in a Private or Hybrid Cloud model. Revenue through Consulting Services.
Resources and Revenue Sharing Models: Identify networking partners, AI and Blockchain partners for SW Development and Integration.
Thus, an opportunity definitely exists for further standardization and development in this space, which this proposed system can help to solve.
Disclaimer: Any views and/or opinions expressed in this post by individual authors or contributors are their personal views and/or opinions and do not necessarily reflect the views and/or opinions of Huawei Technologies.