Using AI-driven Portfolio Optimisation to Consider ESG Preferences in Portfolio Creation


Practising socially responsible investing has become a priority for many clients across the globe as markets are shaken in the wake of climate change and a global pandemic. However, catering towards sustainability preferences can be complicated and time-consuming. Using AI-driven portfolio optimisation can help advisors to factor in a myriad of sustainable preferences effortlessly.


SRI is experiencing an exponential rise

Client interest in socially responsible investing (SRI) has been booming lately, with CNBC reporting that despite the market turmoil caused by the global pandemic, sustainability-focused funds attracted a record amount of capital in the first quarter of 2020. As SRI becomes more and more mainstream, advisors and wealth managers need to keep up with the growing demand to focus portfolios, partially, if not entirely, on sustainable funds and investments.

Despite regulations such as the EU Disclosure Regulation 2019/2088 that introduces disclosure requirements regarding sustainability in the financial services sector, advisors may find it difficult to cater to sustainable portfolios. Not only do they have to navigate complex terminology, but simply being assured of what is and is not sustainable can be a complicated task. According to a recent Morningstar panel, advisors feel challenged in the realm of sustainable investments, with many not feeling comfortable enough to have such a conversation with clients.

Furthermore, current regulations merely require advisors to ask if clients have sustainability preferences. In the future, this will become much more granular, with clients expressing a desire for specific considerations such as green technology to be included in their portfolios. This is something we wish to further develop: the ability to construct an ideal portfolio that can cover all specific client preferences.


AI-driven portfolio optimiser can efficiently tackle ESG preferences

Incorporating these factors smoothly into portfolios is where AI-driven portfolio optimisers shine the brightest. Privé has developed an optimiser framework based on genetic algorithms that is able to consider client preferences, which are measured as fitness factors, for various criteria. Here, fitness factor is a functionality of the Privé optimiser AI GO that is used to evaluate the quality of a specific preference. The aggregation of all fitness factors for each client preference results in the total fitness score of a portfolio. Different fitness factors are used to help define the objectives of the investor and hence the optimization process based on these client’s preferences.

Fitness factors can cover financial key figures such as a return or sharpe ratio goal, a volatility cap, as well as asset classes, FX preferences and many more. Advisors or Portfolio Managers can define preferences and assign weights to the factors that are important to the client. This is also a key element of the INFINITECH project, which has set out to create solutions that can integrate as many client requirements as possible within investment advisory processes.

Sustainability can also be incorporated as an AI GO fitness factor.


Container orchestration technologies at the service of INFINITECH
Figure 1: Selecting your sustainability preference can be easy


Recently, Privé started a collaboration with a leading insurance company on an end-to-end advisory process that utilises parts of Privé´s AI-driven portfolio construction framework. By using the clients’ ESG definitions, it is possible there to incorporate sustainability as a key fitness factor. ESG largely refers to three dimensions of sustainability, namely: Environment, Social, and Governance. Investments made with ESG in mind have to be ecologically neutral or beneficial, be ethical or improve living conditions, and the company itself has to adhere to strict ecological and ethical guidelines. In the optimiser’s fund universe, all funds are tagged and filtered according to their sustainability. Here, sustainability can be determined either by the asset manager or from external data providers. The optimiser allows users to select the percentage weighting of the desired ESG compliance or sustainability grade of a portfolio proposal, and the fitness factor ensures that the desired weighting for ESG compatible investments in the portfolio is reached.


Hyper-personalisation is the future


However, simply allowing clients to select sustainable options is not enough. Hyper-personalisation is an emerging trend in wealth management, with many clients expecting to be able to select, down to the wire, exactly what type of investments are being included in their portfolio. Here at Privé, we expect this to evolve into clients being able to choose which SRI path they wish to pursue, be it excluding fossil fuels or nuclear energy entirely from the portfolio, or avoiding companies and funds with a history of child exploitation or modern slavery.

INFINTECH funding has contributed greatly to our efforts in creating the AI based optimiser framework. Currently, Privé is researching various methods in developing fitness factors with sustainable subtopics in line with the United Nations’ 17 Sustainable Development Goals. These subfactors will, in the future, be incorporated into Privé´s AI-driven optimiser to further bolster our original framework based on further customer preferences.

Managing sustainable and ESG compliant preferences and juggling hyper-personalisation is no easy feat. However, with the help of Privé’s AI-driven optimiser engine, this process can be made virtually effortless and encourage both advisors and clients to pursue SRI. With sustainability a topic that will increase in popularity as the climate change discussion intensifies and becomes more result-driven, the wealth management sector must be ready to meet such demand.
Logo

Are you ready to work with us?
Send your inquiry now

info@infinitech-h2020.eu

Invalid email address.

logo europe This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 856632.
The content reflects only the authors’ views, and the European Commission is not responsible for any use that may be made of the information it contains.