Data and Tools for a Sustainable Circular Economy
If we are to seriously tackle sustainability challenges and the environmental pressure of our production and consumption system, we need to improve data and tools and strongly institutionalize their methods.
This can be done by ensuring statistical offices and institutions do frequent and systematic collection of data useful for the circular economy. This applies both to the macro scale (i.e. the overall global arrangement of consumption and production) and to the micro scale (i.e. the supply chains of specific products or services).
The current data challenges limit our analytical capabilities and the development of tools to aid decision- making reliably and transparently in the private and public sector. In this article, I will first highlight some of the challenges related to data availability and software. This is followed by a brief overview of what is currently being done and concluded with some ideas open to discussion on how this can be solved.
“The current data challenges limit our analytical capabilities and the development of tools to aid decision-making reliably and transparently in the private and public sector”
Data challenges at the macro-level
Environmental macro-economic analysis often relies on Input-Output data. This data is collected by national statistical offices and then compiled in such a way that the structure of the economy is represented. There are many ways to do this, but the current models fail to represent many aspects that are important to the functioning of our production and consumption system.
It is well known for instance that the reporting and representation of some important pillars of our society is unsatisfactory. The most well-known case is perhaps the value created through informal labour such as work provided by spouses. But better representation is needed also for services provided by ecosystems, the value created by remediating polluted areas and the accumulation and processing of disposed products (e.g. material stocks through society).
Databases used for scientific research try to address these issues, but they often struggle with keeping up with this herculean task due to lack of resources. They often rely on occasional national or supranational projects and, as such, are not institutionalized into the working of the official statistical offices. This leads to sporadic updates and insufficient expansion.
We often find:
Low resolution of product categories (I.e. families of products serving similar functions)
Limited data on the physical transactions between industries
Limited data on material stocks and waste created
Disconnect of the models from the relationship with natural capital (e.g. ecosystem services)
The list could go on, but it could be summarized in the need for an expanded and systemic data collection effort. This work needs to start from the need for a greater representation of product categories and include expanded accounting on natural capital with great attention to ecosystem services.
For instance, in the product category classification offered by EUROSTAT under PRODCOM, we see that the classification of products (i.e. commodities and activities) covers around 4000 listed groups. This can be considered a very comprehensive list. Information on the trade of these products is accessible but information about their production inputs, stocks and environmental impacts is very limited as it is only partially collected. This, and other data collected by EUROSTAT, is then used to compile Input-Output tables . However, EUROSTAT Input-Output tables include only 65 products categories under which all the 4000 groups we mentioned before are aggregated. While other databases exist for macro-economic multi-regional analysis that have a more extensive product resolution, they still do not exceed a few hundred product categories and database update is handled on a project by project basis (e.g. by universities and private consultancies).
This is a serious matter as it is unavoidable that a limited representation at the macro scale negatively influences the way we design and implement policies, and the risks we take along the way.
“It is unavoidable that a limited representation at the macro scale negatively influences the way we design and implement policies, and the risks that we take along the way”
Data challenges at the micro-level
The lack of a systematic work at the macro-level can also be seen at the micro-level in data collection concerning products’ supply chains and their impacts. The data collection on the impacts of a specific product is often compiled in Life Cycle Inventories (LCIs). These are the datasets backbones of Life Cycle Analysis (LCA), which shows all inputs and outputs to the economy and environment across the life cycle of a product.
While it needs to be acknowledged that LCI databases have grown substantially through the years, current data collection methods are not well-equipped to face the reality of the incredible amount of stuff we manufacture. For instance, one of the most broadly used LCI databases, Ecoinvent, includes around 3000 unique products (not accounting for process and geographic differences). While this is certainly an impressive number of LCIs that has been created by meticulous scientific research, we know that according to some estimations each year around 30,000 new products enter the global market (this does not include capital goods). And even though only ~5% of these consumer products survive their launch, this is cumulatively still a substantially larger number than the data available.
We are lacking sufficient data to discuss rationally about the socio-economic and supply chain changes required by the transition to a circular economy. This is—in large part—because collection of data for LCIs is notoriously time and labour-intensive and, most of all, entirely voluntary. Companies can freely decide whether to do it or not. It is also important to note that the collection of data along the supply chain of a product is not an easy task. Companies can be faced with a great deal of opacity across their operations, mostly if they take place along international supply chains. Which most do. This means that unless a company adopts strong internal practices for Product Lifetime Management and Product Data Management, data for LCIs cannot be collected systematically.
“We are lacking sufficient data to discuss rationally about the socio-economic and supply chain changes required by the transition to a circular economy”
Because of the voluntary basis and the general reluctancy to provide, collect and store sensitive information that could affect the protection of intellectual property, LCI databases are often maintained by a few private organizations (e.g. Ecoinvent, GaBi, etc) and only limitedly by public institutions. Because of this, we encounter issues of data interoperability due to the lack of enforced common protocols. And it results also in challenges in access to the available data, ultimately creating inequalities in the way these scientific outputs are used by scientists and businesses in different regions of the world.
This also hinders our ability to take preventive measures against potentially environmentally damaging productions as well as monitoring supply chain performances globally and locally. These factors affect business competitiveness and companies’ ability to tackle environmental and circular economy challenges. Companies that want to do better environmentally through the circular economy may not know how impactful their competitors are. On the other hand, those who are being impactful may not have the competitive pressure to curb their harmful activities.
Furthermore, the lack of data limits us in assessing the addition to material stocks related to specific supply chains, which means that we are also limited in addressing products End-Of-Life solutions both from an institutional standpoint but also from a business perspective.
Assessment tools challenges
The relative limited data and lack of institutionalization leads to another issue: it is difficult to build assessment tools for data that is not available or not consistently collected. However, this is not the only factor. The relative novelty of the field also influences the current availability of tools. The concept of the circular economy tries to bring together multiple disciplines into one common framework. Disciplines that carry with them their own established methods and challenges with data, methods, ontologies, and tools. These need to be bridged to ensure that insights from the micro-level can be useful for the macro-level and vice-versa.
If we look at the availability of software for macro-economic scenario analysis, many commercial and non-commercial alternatives do not include data on natural capital, and they are not structured to aid circular economy decision making. They usually do not allow for the comprehensive assessment of circular economy aspects such as changes in material stocks, waste generation and servitization of commodities. Some applications allow for the assessment of physical flows in the economy, but they often lack the possibility to make simulations of plausible alternatives. This means that analysts need to rely on developing their own software for their research questions. There is nothing wrong with that and it will always be the case for certain types of analyses. However, it would be helpful if features for circular economy modelling and assessment would be available in popular software at different level of analysis and that source code was made publicly available to ensure transparency and reproducibility of studies.
“It would be helpful if that source code was made publicly available to ensure transparency and reproducibility of studies”
Furthermore, while researchers and institutions have been at work defining which circularity and sustainability indicators are relevant, fragmentation still exists. At this moment, there is a question of how indicators concerning Sustainable Development Goals (SDGs), planetary boundaries, and circular economy will be integrated in software for sustainability assessment.
Positive developments on data and assessment tools
Many of these challenges have been on the radar of experts. For instance, the European Commission is currently working on strategic “dataspaces”, one of which will be the “European Dataspace for Smart Circular Applications” which should include digital product passports providing “information on a product’s origin, durability, composition, reuse, repair and dismantling possibilities, and end-of-life handling” . Recently, United National Environmental Programme (UNEP) together with a broad coalition of countries and the Life Cycle Initiative launch GLAD, which tries to centralize access to LCIs and it aims to promote interoperability of datasets. The United National Statistical Division (UNSD) is currently working towards a better use of big data to improve data collection.
In this respect scientists are looking into ways to automate data collection methods to increase our analytical capabilities. Some are working on making sense of indicators and tools to facilitate Circular Economy assessment by stakeholders. While others are looking into helping companies' circular operations through Product Lifetime Management.
Companies are also seeing the advantage of increased collection of this information. Some are beginning to use block-chain technologies to track materials across supply chains and help closing supply chain loops and others have built resource passports and waste-to-resource marketplaces. Others are implementing Internet of Things technologies that could help us in monitoring products performances, and where and how they accumulate at their EOL.
Together with the increase in data collection efforts, available tools are being developed for sustainability and circular economy. The Ellen MacArthur Foundation developed Circulytics which aims at supporting companies through circularity measurement. Recently, The Hotspot Analysis Tool for Sustainable Consumption and Production (SCP-HAT) developed by WU Vienna was also launched. The tool allows to compare national and sectoral sustainability performances and raw material consumption. There is interesting work that is done by the Stockholm Environmental Institute in combination with Vizzuality, where they combined supply chain analysis methods with a detailed regionalization of supply chains for various agricultural products and their risk on deforestation and CO2 emissions. Makersite, a commercial online software, is now offering Circular Economy features for Life Cycle Analysis. And there is, of course, the work we do at Leiden University CML, where we offer analysis and modelling of scenarios using Environmentally Extended Input-Output data through the RaMa-Scene and CircuMAT platforms.
Some ideas on how to solve these challenges
After highlighting some of the challenges and developments, I would like to bring forward some proposals open for further discussion and which reconnect to current talks about expanding the role of statistical offices and about data collection.
Data at the macro-level
As mentioned previously, many aspects of society and the circular economy are not sufficiently accounted. The work of statistical offices needs to be expanded into collecting more data on production, consumption, stocks and natural capital and it should be done frequently.
The current system of national accounts could be updated to ensure that the following data is accounted systematically and regularly:
hybrid/physical accounting (I.e. mass, volume and energy content transferred through all economic actors)
material stocks, losses, and dissipation (i.e. how and where materials accumulate through society and the natural environment)
Natural capital accounting both in terms of environmental impacts—which is partly already done—but also on biodiversity (e.g. pollinators, bird species, etc.)
Critical raw materials most of which are used in our technologies and have become a security concern for countries
And accounting concerning planetary boundaries such as the flows of phosphorus and nitrogen compounds
To achieve these points, we will need to step-up data collection efforts through fieldwork but also take full advantage of the latest computer science techniques to fill in data gaps.
“We will need to step-up data collection efforts through fieldwork, but also take full advantage of the latest computer science techniques to fill in data gaps”
Global multi-regional Input-Output tables should be compiled every year using this data and providing a higher level of product detail. Ideally, the product categories would be as many as possible, which should be around 3,142 if we followed the sixth-level of the UN Classification of Product by Activities. This is of course an incredible effort to be made. However, if we reached even the fourth (575 categories) or third level (261 categories) while including a few strategic product categories from the lower levels, that would already be a significant improvement.
Data at the micro-level
LCIs databases also need to be expanded rapidly in a way that respects FAIR (Findable, Accessible, Interoperable and Reusable) principles. We need to make this information publicly available so that decision makers and the public can easily access it, regardless of their geographic location.
Scientists should spend more efforts studying the impacts of already existing supply chains. This should include data mining and machine learning techniques which could automate data collection and speed up our assessment efforts.
Companies with production over certain volumes should have the requirement to communicate their LCI to a centralized LCI database to participate in the market. This database should also show clear connections to circular economy aspects that are typically not included in LCI (e.g. lifetime of a product). A system resulting from the combination of GLAD and European Dataspace for Smart Circular Application could be the starting point of this effort. Intellectual property concerns in such system could be addressed by methods of data anonymization.
“Companies with production over certain volumes should have the requirement to communicate their LCI to a centralised LCI database to participate in the market”
Assessment tools
At last, it is important to revise the models and instruments we use to make predictions to plan societal wellbeing and environmental stewardship. For example, circular economy considerations could be added to commonly used Integrated Assessment Models so that they can be better coupled to global decision-making concerning climate change mitigation. But they also need to be included in less computationally intensive tools for exploration of sustainable alternatives. Tools for macro-economic assessment and LCA, could integrate procedures to assess circular business models and make scenarios using physical information about the economy such as materials stocks.
“It is important to revise the models and instruments we use to make predictions to plan societal wellbeing and environmental stewardship”
Whatever the level of detail, assessment tools should enable users in capturing the multidimensional aspects of circular economy and sustainability, while also allowing for transparent and reliable scenarios for a sustainable future. Which means that clear connections between circularity, SDGs and planetary boundaries should be made obvious to modelers and analysts using these tools.
July 2020
Franco Donati
Franco Donati is a PhD student in Industrial Ecology at Leiden University Institute of Environmental Sciences. His work focuses on the study and development of tools and data for Circular Economy scenario making at the product level (Life Cycle Assessment) as well as the macro-economic level (Environmentally Extended Input-Output Analysis).
The Research Series wishes to offer a thought-provoking and open space for researchers worldwide to present and circulate their research ideas for a collective debate. Any comment, disagreement, embracement, or question from the readers is highly valued and appreciated.
Recommended links for further reading & discovery:
“Circular Economy Indicators: What do they Measure?”, article by Moraga et al. (2019) - open access
“Following-up on opportunities for a circular economy: better data for robust policy-making” , publication by Rietveld et al. (2019), TNO - free download