Discussion Paper
No. 2017-82 | October 06, 2017
Kai Neumann, Carl Anderson and Manfred Denich
Beyond wishful thinking: Explorative Qualitative Modeling (EQM) as a tool for achieving the Sustainable Development Goals (SDGs)
(Published in The Sustainable Development Goals—Assessing interlinkages, trade-offs and synergies for policy design)

Abstract

The UN’s Sustainable Development Goals in their generalized form need to be further reflected in order to identify synergies and trade-offs between their (sub-)targets, and to apply them to concrete nations and regions. Explorative, qualitative cause and effect modeling could serve as a tool for adding crucial factors and enabling a better understanding of the interrelations between the goals, eventually leading to more informed concrete measures better able to cope with their inherent obstacles. This work provides and describes a model that could serve as a template for concrete application. The generalized model already points to some potential ambivalences as well as synergies that can be reflected on using some of the latest theories and concepts from economics and transition research, among other fields. Its first analyses cautiously raise doubts that some possible assumptions behind the original Sustainable Development Goals might overlook some systemic boundaries. For example, an undifferentiated increase of productivity contradicts a lessened environmental impact and need for resources in light of potential planetary boundaries.

JEL Classification:

A19, C19, B49, C38, C69, O10, O19

Links

Cite As

[Please cite the corresponding journal article] Kai Neumann, Carl Anderson, and Manfred Denich (2017). Beyond wishful thinking: Explorative Qualitative Modeling (EQM) as a tool for achieving the Sustainable Development Goals (SDGs). Economics Discussion Papers, No 2017-82, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2017-82


Comments and Questions



Anonymous - Referee report 1
October 06, 2017 - 12:42
The paper contributes to the debate on SDGs, yet• style of language and illustrations needs to be improved to be in line with scientific journal standards,• the application (including the participatory method) should be at the center of the paper rather than the description of model features potentially available to the user, and• the pros & cons of the modelling approach in the context of SDG tradeoffs/synergies need to be discussed somewhere, currently the conclusion section sounds somewhat to enthusiastically optimistic

Anonymous - Thank you very much for the suggestions
November 20, 2017 - 15:46
We will have an additional look at the chosen language and adjust accordingly. With regard to the illustrations: they are not to present any concrete result. Rather they show the tool and the fact that the tool shows factors and matrices. It is not crucial to read the actual texts within the screenshots and zooming the pdf it is possible to read most of it. We will, however, improve the quality of select images and consider your suggestion. The paper shows indeed both the qualitative model on the SDG targets and their potential interconnections that need to be adapted to a concrete region as well as the participatory, explorative qualitative modeling that can be used to add crucial soft and hard factors to a concrete challenge. If it was only the latter the example would be too specific for the purpose of the paper, as this addition was more to show the way in which the model can and should be further applied and amended at a sub-global level. We will attempt to better frame the inclusion of this example within the overall contribution and further elaborate where possible. We agree with reference to the pros and cons of the modelling approach, this can be made more balanced and relate to the numerous general obstacles one should be aware of when attempting such an analysis. Some of these potential barriers therefore not only in the context of the SDGs but also the modelling approach need to be added. (the team of authors)

Anonymous - Referee report 2
November 09, 2017 - 11:43
(i) Is the contribution of the paper potentially significant? The papers comes to meet a pressing need in comprising the overwhelming diversity of SDG targets and their complex relations in tangible tools for use by decision makers. In this respect the contribution of the paper is of great significance. Furthermore, there seems to be scope for improving the analysis done in order to make the paper a more solid work. The significance of the paper would increase, in my opinion, if the authors link the selected model excerpts of figs 5 and 7 to concrete case studies (in peer review literature or good reports from authoritative institutions like UN, World Bank, GIZ, etc). In particular they should reflect on how their qualitative finding map to more quantitative aspects and socio-economic settings on the ground. (ii) Is the analysis correct? It is as correct as it can be given its qualitative nature but some important details are necessary to work on. As it is not the objective of the authors in this paper to draw quantitative judgements on the feedbacks between SDG targets it is not feasible to state if their analysis is "correct". Depending on who you ask or in which socio-economic context you are in the relation between target X and Y can differ. While in some countries X might have lead to less Y, in other the opposite might have been true. In the paper "most connections represent arguments from the authors" "based on their acquired knowledge and commonly accepted relations." Without an ampler participation of experts informing on the relations between targets the model stands only on the shoulders of the authors. It would be advisable therefore than in order to enhance the "correctness" of the analysis, the authors include more expert-base perspectives, including the ones that will prove contradictory. From the provided description the modelling tool used does allow for broader participation of opinions to be easily captured. Furthermore, the authors are further developing the model using specific information in workshops and therefore they have the connections to engage on a broader discussion. Another point that I find strange, or miss understood, is why the relation between target X and Y has to be either a + (plus) or a - (minus). Why is there no scope for a neutral relation 0 (zero). The authors also seem to assume for the moment on a "universality" of relations. independent of which country you are in target X is always related to target Y in the suggested way. This defeats one of the big potentials of the SDG's, that is, to learn from best practices in countries or regions that achieved already target X without negatively influencing the achievement of target Y. In summary: more perspectives on the relations beyond those of the authors is required.

Kai Neumann - Thanks for your insightful feedback!
November 20, 2017 - 15:47
Incorporating more expert literature would perhaps provide the reader with more resources for further study and help foment thought and conversation around the topic, possibly also providing an idea of potential tradeoffs and points to consider when creating such a model. Without this inclusion we feel it is still a tool (the SDG targets connected in a freeware) and a methodology (participatory explorative qualitative modeling) to gain insights. The idea is that others take the generalized model and perhaps adapt it in the way you have suggested. That being said, we acknowledge that an incorporation and reflection on key literature or reports as you have suggested would improve the work and we would be willing to include this.We should make it more evident in the paper that the original targets as provided with the factor descriptions in the model are still somewhat vague and that indeed in the context of a concrete region one should be quite specific. One suggestion to be included is to ask how a factor could be measured. Sometimes one will end with numerous specific factors instead of the current target factors. The current model just helps to connect the targets - it is not a concrete example for a concrete region. With regard to the model being based on our own work; as previously stated, we have to make it clearer that this is not a concrete region and therefore an attempt at the global level even with experts would be ill-informed given, as you explain, the heterogeneous nature in terms of SDG interactions. Perhaps we can make this more obvious or our contribution more transparent. We will attempt to incorporate more expert based opinions but they will inevitably remain more of discussion points rather than warrant statements regarding “correctness” of the model, due to inherent subjectivity and contextuality. The model from the concrete workshop was included to show how a specific topic can be integrated into the context of the SDGs and drew on work from a project with a different explicit aim. Yet, the example wasn’t elaborated as this would have required weighting the interconnections between the SDG targets to fit the case of Ghana. Regarding the comment about only either positive or negative connections and no “neutral” connections - cause and effect modeling does indeed require to decide whether there is a plus or minus in order to analyze feedback loops that could be reinforcing or balancing. However, one can either simply not connect the factors (the assumption here would be that there is no influence) or connect the two factors with a bi-directional connection. In other words say Factor 1 leads to more of Factor two AND Factor 2 leads to more of factor 1 (or some combination therefore). It was certainly not our intention to argue based on any universality. Obviously, we have to make it clearer that it is just a template one needs to continue to adapt to a concrete region. While we do agree that sharing best practices is crucial, one shouldn’t assume that the direct connections between the targets are the same in every country and that lessons can be extrapolated. We would encourage the more specific focused models created to be shared and open source on the online platform provided in the article to foster learning, along with qualitative descriptions of lessons learned and how they are captured in the models or modeling process. We will include this more explicitly in the article. We will attempt to incorporate more perspectives for fostering thought and reflection but feel that the contribution of the paper is in fact to provide the outline, approach and considerations for others in order to capture perspectives which would be relevant to their respective models rather than provide generalized prescriptions.

Anonymous - Incorporation of empirical data to explore inter-linkages
December 01, 2017 - 10:00
This is an interesting study that provide exploration qualitative modeling tool for visualizing SDG interactions. However, this study may incorporate empirical data to explore inter-linkages. Currently, the assumptions made on direction, strength, and intensity of interaction among the SDG targets is not clear and transparent. Following is a recently published paper on SDG interactions using empirical analysis. A Systematic Study of Sustainable Development Goal (SDG) Interactions (http://onlinelibrary.wiley.com/doi/10.1002/2017EF000632/full)

Kai Neumann - Correlation vs. Causation
December 08, 2017 - 09:29 | Author's Homepage
Thank you very much for the very interesting paper! We are discussing it currently. The paper itself notes that correlation doesn't need to mean correlation and our paper offers a methodology coming with a tool to closer examine through which processes the targets are actually linked. The more detailed interlinkages, then, need to be based on something. Best would be scientific 'proof', followed by stakeholder's mental modeling, followed by educated guesses, followed by mere guesses. Data could be used for both, to verify or to falsify the interlinkages, though in both cases there still could be other reasons that are simply not included, yet. So, looking into the future science could only offer abductive logic and neither the experiences from the past (conclusions from data) nor the assumptions within a cause and effect model can be more than an preparation for a possible development. If data falsifies a model the modeler needs to add some factors to his or her model in order to explain the different outcome from data or the model is simply wrong. By the way: to build a model from data bears to risk to draw direct connections that are actually already existing indirect connections. I think we will clarify these aspects in our paper and leave both the combination of the two approaches and the application to a concrete region to another paper or project. The author from the other paper has contacted us so we are eager to further develop these approaches.