With a total shutdown of certain sectors of activity, it was not surprising that in June, leading economic forecaster EY Item Club predicted that GDP would shrink by 8% this year. Though the drop was not as dramatic as expected, the Office for National Statistics reported that GDP in June was still a sixth below its level in February, before the virus struck. 

The numbers are indicative of a lack of visibility on the timing and extent of economic recovery during this pandemic, and as well as market volatility. In these unprecedented times, it’s also become very difficult for banks to judge their assets and to assess the credit quality of both individuals and companies. 

Many banks are already likely struggling to understand and anticipate market fluctuations, and without alternative data, AI, ML and agility and reduced time-to-deployment in modelling, the road to recovery will be even more challenging. The idea of environmental and socially responsible recovery, which has become pronounced in the pandemic landscape, adds an extra dimension of complexity that banks must look to overcome. 

Enrich the range of data collected

In this age of uncertainty, banks and asset managers are being forced to re-think their approaches by deploying new rules and integrating new information. The traditional — and often static  — data that has been used to access the financial health of individuals and companies is no longer good enough. Financial results, debt, market share, international presence and diversification, benchmark, partnership and competition data are not going to fuel the rapid, dynamic decisions banks need to make easily in today’s landscape. 

Banks must add new alternative data such as news streams, geolocation data, satellite images, studies, reports and even scientific articles. All of this data, whether or not it is structured and processed by AI algorithms, makes it possible to collect different and more immediate perspectives on the health, risks and developmental potential of each account. 

However, in the current context of pandemic and climate change, recovery will have to be responsible. The crisis has shown that new paradigms are possible, and that bending the curve of CO2 emissions is within our reach with the proper alignment. Amongst them, banks and asset managers have a key role to play in integrating these environmental and public health considerations as widely as possible in portfolio management as well as in risk assessment and corporate financing. 

The very essence of Sustainability and Corporate Social Responsibility (CSR) data is complex, as it covers concepts as diverse as the impact of the company's activity on human health, its respect for the environment, its non-discriminatory policy towards minorities, or its virtuous actions towards society. Varied in content and format, this information is difficult to collect and use. 

AI and ML are making it possible to extract meaning from masses of heterogeneous CSR data without obvious correlations, and thanks to this data and these tools, banks and asset managers are able to structure products that are more environmentally friendly and socially responsible. Moreover, AI and ML will become increasingly crucial for banking players and asset managers in order to evaluate companies across this broad spectrum of dimensions and structure their offers. 

That being said, AI is not a magic bullet: its effective integration into the processes of financial institutions is demonstrably taking time to prove. The learning curve must be supported by constant collaboration between data scientists and banking experts, with the aim of co-construction and acceleration. New data sources must be integrated in order to make gains in modeling accuracy, leading to making better decisions about lending to individuals or businesses, while visualising the impact of actions on books of business. 

It’s a step in the right direction to better understanding today’s economic uncertainty, while also rising to the ongoing challenge of climate change, and if banks can work with this new data, they will be well positioned to be agile and active in driving an environmental and socially responsible recovery.