A SDGs network analysis to enhance coherent evidence based policy: For SDG, the devil is in the interactions

It is widely understood that in order to achieve the 2030 agenda we need to implement evidence-based policymaking, but recently It has started to become clear that it is also key to design coherent policies, and for this, we need to incorporate a complexity approach to take into account SDGs interactions. We construct an interaction (Mutual Information) network using data from SDGs progress in most important metropolitan areas in México. We perform a node ranking analysis and compare the results with a theoretical network. We show that empirical and theoretical networks have a different focus, showing that SDG needs at least different policies for urban and rural areas. We also analyzed the effect of individual (ignoring interactions) SDG achievement using a Bayesian Network. We show that in general monolithic SDG policies translate to a higher probability of low progress. We also made some comments related to the current monolithic general focusing on Poverty fighting by the Mexican government in contrast with what available data suggest would be better: Decent Work and Economic Growth but with a Responsible Consumption and Production. Then investing in education and scientific research in order to advance in Industry, Innovation, and Infrastructure, which requires a strong law and justice procuration system that enables peace, justice, and strong institutions.