Feb 2023
Sustainable finance - do we really need more data?
Arithmomania is a type of Obsessive Compulsive Disorder (OCD) characterised by the inability to stop counting objects and making mathematical calculations. It can be disabling and overwhelming for someone who suffers with the condition. When severe it impedes work completion and it creates serious anxiety. The latter occurs as the mind is busy with counting, so letting social interactions aside and not being given sufficient attention.
Researching for this blog on data and its use in sustainable finance, a new type of organisational arithmomania came to mind. In my experience, working with the financial services industry and accountants, obsession with data has never been greater. Part of the reason is the financial value which our personal data has, for example, feeding into social media algorithms to target advertising. In 2017, the World Economic Forum estimated the global data economy was valued at $3 trillion. Companies which make use of data, such as Facebook and Google, now have some of the highest stock market valuations in the world.
Sadly the trend of cyber attacks to obtain personal data and records is also increasing. This has the potential to cause significant damage to the economy if the institutions which are hacked are ill prepared. Added to this, anecdotally consumers are often asked to share more and more personal details via websites in order to fully participate digitally. It is easy to see how this can lead to theories around the potential for a surveillance state and the possibility of being locked out of services due to what is shared. Having said that, some people say, if you have nothing to hide then this is not an issue. Edward Snowden may disagree.
ESG-related data
When it comes to ESG (Environmental, Social, Governance) related data, organisational arithmomania can manifest itself in different ways. Are any of the following questions familiar?
What standards were used to prepare the data?
What is the coverage and scope of the data?
Is the data comparable and complete?
Has the data been reconciled to the financial/customer records?
Who reviewed/signed off the data?
As an accountant, for financial and regulatory reporting these are often business-as-usual questions. International accounting standards have been around since the 1970s and audit checklists have many of the questions above. However, maybe some of those questions do not help so much when it comes to newer ESG-related data types?
For example, many companies report the consolidated results of their employee engagement surveys. This includes % responses to questions such as “Would I recommend this as a place to work?” on a scale.
Even with written guidance, people will respond based on their own perception. The responses could vary depending on the simple interactions on the day of completing the survey. A smart former colleague of mine at the bank once said, it is similar to being asked “how much do I love my partner?” - it could really vary depending on the circumstances of that specific day! There are also unintended consequences around managers requiring employees to fill in such surveys with positive responses under duress. Culture also may influence the answer, for example, from a very expressive positive / over enthusiastic culture to a more disillusioned and self-deprecating one.
Furthermore, there are potential for other unintended consequences when collecting other types of ESG-related data. For example, employee information such as gender, ethnicity, sexual orientation, social class, etc. This is often private information which could be used to discriminate, and/or stigmatize individuals within an organisation. There are benefits to collecting, understanding and acting upon this data. However, it could also lead to a culture of preparing reports ticking boxes on the basis of diversity, rather than working on impactful inclusive actions.
So, why?
There is a perception that data is the truth, that it is ‘neutral’ and that skilled managers will act and make data-led decisions. However, all data is ultimately crafted in some way by humans. Even if machines complete a calculation, the code is written by a human. When data is presented as information, it often comes with human bias. The words and discussion around data being presented can easily influence decisions more than the underlying data itself. Also suppliers sometimes have their own self-interest - i.e. selling more data and standards - as a motivation in the quest for more data.
Jessup: You want answers?!
Kaffee: I want the truth!
Jessup: You can't handle the truth! Son, we live in a world that has walls, and those walls have to be guarded by men with guns.
- Aaron Sorkin - A Few Good Men
Perhaps data is a type of comfort blanket. It is reassuring, as we perceive we take ‘data-led decisions’ which are free from human emotion. For example, regarding sustainable finance, sometimes the data sought is used to convince a financial institution acting in a sustainable way will not negatively impact their financial results. This approach can miss the easily miss the bigger picture around the purpose of acting sustainably.
Everyday humans take plenty of decisions without extensive data-sets, KPIs (Key Performance Indicators) and metrics. There are some things we know intuitively and act upon with purpose. For example, taking care of a loved one, laughing at a joke or crying while watching a film. In some cases, using data has the potential to limit our decision making as we ignore qualitative aspects of life and business which are harder to measure. In business setting this includes professional judgement and the use of written reports.
Pursuit of good enough data quality can also create internal interia inside an organisation (sometimes on purpose). For example, calculating and recalculating without getting to any decision. Obtaining data and reporting it is not the end in itself. This can be observed with greenhouse-gas accounting. It can be challenging to calculate a financial institution's scope 3 carbon emissions based on its lending book to the highest “quality” - often due to lack of data from the underlying customers. The PCAF (Partnership for Carbon Accounting) standard allows for some options in quality, but there are nevertheless potential trade offs. The same as also be said for Paris Agreement alignment data.
For example, before implementing a new policy on restricting oil and gas lending at a bank: Is it better to wait for an impacted company to self-report their data, or should the financial institution immediately start to implement based on emissions estimates using proxies combined with scientific research? More data of a better quality isn’t necessarily better. Estimated data combined with sound judgment can certainly be enough to make a decision.
A few questions which can be helpful to ask in these types of situations:
What is the purpose of this data?
What type of data do I need specifically for this decision?
How ‘accurate’ (of what quality) does the data need to be to make this decision?
What are the consequences of inaction while waiting for the data?
What are the potential unintended consequences (often social) of using this data?
Conclusion
On a final note, data privacy will continue to be a key consumer issue. In the pursuit of more ESG-related data it is worth also considering the ‘social’ trade-offs. For example, as customers are we comfortable with sharing more personal information and potentially being monitored by corporations? Finally, with the rise of ChatGPT and other artificial intelligence, perhaps it is time to further consider qualitative information combined with human judgment for our decisions. Similar to exposure therapy, one of the treatments for Arithmomania, facing our fears and adjusting our mindset may help alleviate our organisational obsession with data.
Rebecca Self is the Founder and Managing Director of Seawolf Sustainability Consulting, more information is available via www.seawolfsustain.com and Rebecca can be contacted directly via seawolfsustain@gmail.com.