*HOW DOES IT WORK?*

**Figure 7** provides an overview of the overall process flow and how the relationships between the different components.

At organizational level, the company would need to select quantitative social impact metrics to be included on its report. Those KPIs should be preferably related to the accomplishment of SDGs and aligned with EU reporting standards in a format that can be processed and is interoperable with the Supervisory Board. Specifically, in the case of our solution we have selected html format enriched with

**Figure 6.** *Building blocks of the proposed approach.*

*Social Impact Returns. Filling the Finance Gap with Data Value DOI: http://dx.doi.org/10.5772/intechopen.97407*

**Figure 7.** *Process flow. Relationship between components.*

XBRL tagged metrics, as an extension of the standard that is already used in financial reporting. The reason for this selection is to ease company's adoption of something they are already familiar with, while lowering barriers from the Supervisory Board to introduce new standards.

As we are in a Regulatory context the need for ownership, traceability and nonrepudiation is imperative. Additionally, the technology must be able to address the three pillars of Basel II Accord (Data, Certification and Transparency). Considering its intrinsic characteristics, we opted for Blockchain to provision the Social Impact Reports. Blockchain, beyond its ability to prove trust and immutability of data, it provided another added value: current absence of a Regulatory Body that feels responsible for certifying social impact reports. In a traditional approach, this need would have reflected in additional resources and staff, which we have been able to optimize by means of technology.

As companies are complying with regulatory requirements, they need to receive something in exchange. In our model, we will reward companies with tokens for complying with the regulation. But beyond regulatory compliance we want to incentive companies' alliances to contribute towards the 2030 Agenda SDGs. For this purpose, in our model, companies will also receive rewards for meeting the United Nations' thresholds to attain the expected results.

Results will be accessible on a Dashboard to monitor performance and ensure transparency and fairness of the reporting. But there is something still missing to guarantee the sustainability of the model: people involvement; the qualitative dimensions that provides feedback to the overall model. Based on gamification and AI we generate user engagement to contribute to improve the indicators and benefit from the social impact returns. **Figure 7** shows the overall process flow and the relationships between the different components while **Table 1** shows the value proposition that the model brings for each of the stakeholders.

#### **2.6 PoC: balancing policy measures vs. economic activity**

#### *The Challenge.*

Find the balance between mitigating policy measures and maintaining economic activity.

The Climate Act calls for a 49% reduction in greenhouse gas emissions by 2030, compared to 1990 levels, and a 95% reduction by 2050. The National Climate Agreement contains agreements with the sectors on what they will do to help achieve these climate goals.


**Table 1.**

*Value proposition.*

*Participating sectors.*

Build environment, Electricity, Traffic and transport, and agriculture and land use.

*Proposed indicators* (**Tables 2** and **3**).

*Methodology and approach.*

Our proposed solution is inspired by duality and the concept of system of systems [31]. Duality is twofold and implies:



**Table 2.**

*Proposed Indicators for GHG (green house gas emissions).*


**Table 3.**

*Proposed indicators for energy.*

*Social Impact Returns. Filling the Finance Gap with Data Value DOI: http://dx.doi.org/10.5772/intechopen.97407*

Due to its capability of learning complex structures in large datasets, deep learning has been applied to many problems in financial markets and sustainability, such as analysis and data modeling to design strategies for investment and trading, prediction of prices, identification of market trends and customer behavior and even maximizing profits and returns. There are even examples of applications of AI algorithms to analyze robotic behavior in Smart cities and to understand the impact of news and information on human decisions and arbitrage [32–45].

The quantitative & qualitative factors both reflect on the solution framework. Beyond the quantitative level based on pure mathematical methods, it incorporates human attributes and capabilities of neurons and human learning. Narrowed to practice, our methodology combines a semi-supervised learning method with Generative Adversarial Imitation (GAIL) and Recurrent Neural Networks (RNN). The figure below illustrates the operationalized framework (**Figure 8**).

The framework is structured into three parts: (1) environment, (2) RNN and (3) GAIL. The environment is a virtual place in which we emulate how the

#### **Figure 8.**

*Operationalized effective management framework to balance policy measures of climate change vs. economic activity.*

environment is changing, the impact of climate change and the status of actions and initiatives, etc. Such emulation of the reality helps us practice and identify how Policies can influence and improve the economy. To simulate a realistic market, the environment provides its status (environment state) and the portfolio of climate actions (actions state). The RNN acts as an expert trajectory generator. It produces expert trajectories from raw data (in our problem, training data). Two types of data sources are used in our method: synthetic strategy and real monitored data. During the process of GAIL, we also provide data enhancement to overcome the defects in real data and at this stage we incorporate the gamification factor providing rewards for each state and action.

The actors where this PoC is framed play a crucial role due to their high sensitivity on such a matter as Sustainability and Climate Change, that can have a huge impact on people, the environment and the overarching economic system.

Due to the volatility of datasets, the information from the latest 6-months is generally outdated. Since we are combining privately held and public data to monitor the economic impact of climate change policies in a timely manner, these need to be aligned. Therefore, for the purpose of obtaining relevant results, our approach suggests taking week or few months timeframes, rather than many months or years.

*Leveraging on Big Data.*

The need to combine different types of data and imitate human learning to excel at decision making, demands putting Big Data at the core. It enables to analyze, extract information in a systematic way and deal with large and complex data sets that are too large or complex to be dealt with by traditional data-processing applications and software.

Privately held data is obtained from mobile phone data, internal company engagement surveys, NPS, satellite imaging, while Public data is sourced from National Statistics office, National banks, fiscal studies.

Based on Big Data platforms we are not only able to cope with data with many fields (columns), which offer greater statistical power, but also avoid leading to false discovery rates which are often associated to data with higher complexity (more attributes or columns).

*Algorithm.*

For the gamification module we applied the Loyalty Program Liabilities and Point Values algorithm


## *Distributed Ledger Technologies – why are they important?*

This PoC tackles the Regulatory environment which implies traceability, nonrepudiation and ownership of the results. As of today, in the same way that there are clear responsible Institutions for Financial Reporting Supervision, there is no Organism in charge for non-financial reporting.

*Social Impact Returns. Filling the Finance Gap with Data Value DOI: http://dx.doi.org/10.5772/intechopen.97407*

In absence of this figure, a solution is to rely on Technology: Blockchain. This technology not only provides the necessary principles of traceability, non-repudiation and ownership stated above, but also have proven to be the only alternative to deal with cases that need to combine Regulation and Economic factors.

In such sensitive context where information needs to be immutable, but at the same time there is no one institution responsible for ensuring this, our proposed solution is that each of the agents are responsible for their own information. All in all Blockchain, due to its intrinsic nature, will play the global role that is yet officially unassigned.

#### **3. Results and discussion**

The table below shows a simulation to estimate the social impact of the proposed model (**Table 4**).

With a correlation of 0.87 the model has proven potential to drive social impact returns at SME level, large corporations or country level. The key for the success of such framework is citizen adoption and engagement. All in all, since the model has been developed looking to universal global reporting standards (GRI) and traceability guaranteed by Distributed Ledger Technologies (DLT) it could be extended and tested Worldwide.

In this study, we examined the contribution of non-financial risks to society. When asking companies what is their social impact and the return of their sustainable investments, we often meet a silence. For the first time, with this model, a business that needs to answer this question the next time will be able to provide a quantitative metric. For instance, based on our calculations, a citizen living in a country of an advanced economy, assuming an adoption rate of 65% within a country, could contribute to reduce CHG emissions by 0,97 t during the next 10 years, which would mean that if all countries followed the same example, the objective set by the Climate Act (49% reduction of CHG emissions by 2030) would be feasible to achieve.

It's worth outlining that the backbone of this model is not only the maths and rationality behind, but also adoption and commitment towards a common specific goal. It is considered that for this model to work, a key prerequisite must be satisfied, namely having a joint/compatible goal or problem to solve materialized in a specific metric or KPI everyone understands (it is not enough that parties have their own individual goals and track them in a non-standard way).

It is hoped that this introduction to a new way to measure returns, complementary to traditional finance, will create reflection and commitment to a greater sustainable sensitivity when businesses and event citizens consider how their change of behavior may affect other people, the planet and profits.


#### **Table 4.**

*Simulation. Social impact returns.*
