From forecasting to foresighting: to actively build a better world

06/10/2023

During the coronavirus pandemic, we had to observe that governments and many companies still and mostly develop their strategic decisions based on gut feeling. It has led to poor decision-making processes, in the further loop to a minimisation and even a denial of the crises or shocks that have followed. Without data and without facts, the broader picture is lost. Moreover, without data or facts producing insights in the current situation – the ‘now’-moment’, it is impossible to develop a forecast or to build scenarios embedding the future. Nevertheless: that’s what it's about. If we want to anticipate and steer our future, we will have to think in scenarios.

In the seldom occurrence where we use facts and figures or - even better - developed data-insights to prepare ourselves for a next crisis, we always seem basing our measures on previous experiences. However, a crisis occurs to us as a crisis exactly because it has no precedents. If we, as a society, really wish to steer our future pro-actively and thus control or even prevent worst cases, we will need to think in terms of scenarios. In doing so, we necessarily need to understand deeply the ‘now’ moment and use these insights as a start to work out estimations of what might happen afterwards. Acting that way, we will soon discover that what might occur usually will not immediately correspond to what we had in mind before such an exercise. Further, once we considered risks and opportunities – tangible as well as intangible - and decide to steer the situation in a specific direction, we always might and even will experience unexpected factors that can and will make our future situation deviate from what we want it to be. Therefore, if what we hope for doesn't work out immediately, we will have to repeat the exercise starting from the new ‘now’ moment. Again and again.

A crisis occurs to us as a crisis exactly because it has no precedents.

A complex system

Data has created the illusion of linear predictability. Past findings, coupled with the present moment, allow us to look in a straight line to the future. For certain subfields and very specific cases such as bankruptcy forecasting and fraud-prevention, this actually works quite well on a short term base. Algorithmic models are able to predict failure within a year or a few, but overlooking what will happen within a decade clearly becomes a problem. In a social context, it is therefore completely pointless and impossible to predict where we will be in fifty years.

All the more so because we long time mistakenly assumed that we are rational beings and therefore take rational decisions, while in fact mankind is often emotionally driven. We are overwhelmed by unexpected events, which keep sending us in a new direction. All those rational and emotional elements form complex situations for our systems.

Scenarios

Working with tangible data to develop predictions means that you have concrete, measurable data (e.g. financial data, payment behaviour, etc.) on which you build conclusions forecasting a measurable future behaviour. For example, you can obtain information about the solvency of a company from financial data. The challenge of these mostly linear models, such as prediction models for bankruptcy, is that they measure the problem, but never indicate the solution. Moreover: there’s no such thing as ‘a solution’, since future events might influence the outcomes…. So, we indeed need scenario-building, which is quite different from a prediction.

To start developing scenarios, and thus to asses risks and opportunities to be able to propose different paths of choice, a combination of data, artificial intelligence and human knowledge and emotion is required. The combination allows for two levels of data usage.

First, data that is immediately identifiable, for example solvency, scores, etc. You work with measurable data to achieve measurable results.
In addition, one can also find ‘intangible’ behavioural patterns based on tangible data, meaning that one extracts information from data that, at first sight, has absolutely nothing to do with the information one’s searching for. The datasets of GraydonCreditsafe do not contain information about drug offenses or other crimes. Nevertheless, based of financial data, fleet data and many other but tangible data sources, we manage to indicate how likely a company is to exhibit criminal behaviour. Or in which companies burn-out-development is at higher risk.

Especially the second level gives us the opportunity to go in a completely different direction. It comes down to converting non-measurable elements into measurable elements. The examples GraydonCreditsafe discovered – many times in close collaboration with academics and universities - are surprising to say the least.

Surprising examples

Belgium is an SME country. Although web developers are constructing most of the SME-websites and copywriters many times revise the texts, the entrepreneur still has a strong influence on the final look and feel, and the language used. We found that the language (use of specific words and use of grammar) used on the company-website gives clues towards defining the management style of an entrepreneur. For this, academic collaboration provided a default group of psychologically tested entrepreneurs at the one hand, defining eight typologies evolving from participation to dictatorship. At the other hand, we developed massive web scraping in search for mathematically definable grammar patterns. By crossing these results with other elements such as financial health or payment behaviour, we became able to designate certain companies as dictatorial. Indeed, dictatorial companies do use different language patterns, are often healthy, but they are not necessarily the best payers. They are also characterised by a high turnover of staff. Moreover, once we connected the management style with e.g. lifecycle and activity sector, we also found strong correlations with the innovative strength of a company.

In another study we showed that there is a decent correlation between companies with a high degree of gender equality within the company-board and shock resilience. Another well-known study by McKinsey found that companies, which staff if composed more gender-equal tend to be more innovative. A look at the board of directors and the social balance sheet resulting in tangible data and conclusions about gender-evolutions within a company, combined with their position within a region and an activity sector, enables us to express these findings in a concrete figure. That in turn is one of the indicators that we use in combination with other elements to map out, for example, the innovative strength or the change capacity of a company.

Europe chooses its path

This approach is becoming increasingly important in determining the value of a company. There is the financial value, for which there are all kinds of well-described calculations. But more and more often 'value' is viewed in a much broader sense. After all, value is also found in reputation, innovative strength, flexibility, etc. These are all elements that indeed seem not measurable at first glance.

However, anyone who wants to apply knowledge, facts & figures in social management will end up in an even more complex whole.

Because everything is divided into fields (sociology, law, economics, etc.) that seldom communicate with each other.
Because future-oriented measures are taken in the disciplines that are independent of each other and do not take into account the measures that are dictated by other disciplines.

The question is how we can work with that complexity. Because Europe has chosen a clear direction, in which it wants to see our future society evolve. A choice that clearly sets itself apart from other continents. Everyone knows that Europe wants a climate neutral society by 2050. What people know much less is that in addition to the E (environmental) in the ESG story, there is also such a thing as a social aspiration (Social) and good governance (Governance). Europe has a long-term vision on job quality, gender equality, the way companies are run, etc. And that's not just any political nonsense.

Europe has a long-term vision and that's not just any political nonsense.

Difference based on ESG values

After all, the Corporate Sustainability Reporting Directive (CSRD) has been in force since 1 April last year. Since then, banks are required to report on the ESG values of their largest clients. But it doesn't stop there. From 1 January, 2025, every company will have to include in its annual accounts the extent to which it is engaged in ESG. Meaning they will have to start he exercise in 2024. In addition, it will also have to report to what extent the stakeholders (all customers and suppliers) are involved in this. And with reporting on fiscal year 2024 in 2025, it's high time companies put this on their agenda. Something that has not yet sunk in, while there is nevertheless an entire process involved.

In that context, comprehensive overviews of the stakeholder population are indispensable. Our future path goals on stakeholders value instead of just shareholders value. In addition, it all this will have an immediate impact on the way a company is viewed. Initially, banks, but also investment companies, already not only look at the financial health of a loan or an investment, but also at the ESG values. Companies that score low will no longer get credit or will be granted loans at much higher cost. In time, there will effectively be a huge difference between companies based on their ESG values.

The financial manager is perfectly placed for a leadership role

Incidentally, it is not just about the assessment by the banks. The current generation of young people is much more sensitive to ESG. In addition to the legal evolution, there is therefore also an evolution in the way of thinking of our society. And as abstract as ESG may sound, it will have a direct impact on the financial manager. Until recently, they were solely concerned with risk management. Can I trust the customer? Today their task is much more extensive and includes e.g. fraud-assessment. And in the short term, they will have to ask themselves whether a company not only is creditworthy, but also whether it is e.g. economically sustainable and ESG-proof. The one will be inextricably linked to the other. It also puts the financial manager who is involved in these matters in a perfect position to take on a leading role in this transition.

The fact is that at this stage it is not yet entirely clear what exactly Europe expects in practice. That makes it all even more complex. Banks are currently looking for data in an almost blind panic, but they do not get any further than the bundling of a number of individual fragments.

From assessments to a complete overview

To cope with this, they are currently sending out massive assessments trough questionnaires. Lengthy questionnaires that the recipients – for a fee – have to fill in with good conscience. It results in a cluttered and perhaps subjectively completed pile of individual cases, where greenwashing is plausible, while that total overview is so important to be able to draw up a strategy. A concern for companies, banks and governments, because that overview cannot be obtained on the basis of assessments.

GraydonCreditsafe works hard on ways to make elements that cannot be measured at first sight (intangibles), nevertheless measurable (tangibles) and thus be able to establish additional cross connections. It should lead to a number of elements with which we can indeed measure something at the level of E (environmental), at the level of S (social) and at the level of G (governance). And that should then provide a picture of the situation of the entire Belgian population, but also of customer and supplier portfolios.

The human over the algorithm

As a next step, very concrete support elements can also be built from there. Today we live in a producer-consumer system, while Europe is very strongly promoting systemic partnerships. In doing so, we must evolve from producer-consumer systems to owner-producing units or ‘commons’, in which companies are appreciated trough their particular identity to work better together and reinforce each other.

We must evolve from producer-consumer systems to owner-producing units or ‘commons’, in which companies work better together and reinforce each other.

An important factor here is that data and algorithms, and the conclusions from them, must lead to different paths (scenarios) that can lead to multiple path solutions to answer future risks. However, the final choice must always lie with the person in society. The human chooses its way: it is the human who chooses its future, not the algorithm. To better understand possible choices one can use that data and those algorithms. Firstly, to understand our current situation much better on the basis of facts & figures. And second, to see what would happen if I apply scenario one? And what would happen if I apply scenario two? And scenario three? etc. To better map out possible dangers and opportunities and to make socially responsible choices based on this. So, one does not make a short-term forecast, but a long-term foresight based on scenarios.