The ECA-method for Identifying Sensitive Reactions within an IO Context

The paper introduces an alternative method to determine "sensitivity' within an input-output-context and makes the most sensitive connections visible in an overall graph. The method called ECA (Elasticity Coefficient Analysis) tries-like the Important Coefficient Analysis (ICA)-to derive structures from a certain property of the Leontief inverse. As the core of the method is elasticity-i.e. taking the relative reaction of the element bij of the inverse as a measure of response to a 1% change of the input coefficient aij-it describes the dynamic potential of certain sectors on possible changes of technical coefficients. In the second part of the paper, comparisons are made between the results of MFA (Minimal Flow Analysis), ICA and ECA upon using the same table for analysis. ICA and ECA to some extent show similarities but also some differences. Both contrast with MFA with respect to the very basis of the approach that enforces different interpretations of the results. Finally, some hypotheses on the differences and similarities are set up that could give a perspective for the further use of the different methods.