Geometric composite indicators with compromise Benefit-of-the-Doubt weights

•Proposes compromise indirect geometric Benefit-of-the-Doubt composite indicator model.•Optimistic and pessimistic weights are estimated in a hypothetical, best-case model.•An inter-temporal decomposition identifies multiple underlying policy developments.•A Bortkiewicz decomposition accounts for degree of unbalance in policy portfolio mix.•Empirically applied on EU Member States social inclusion performances.

This paper builds on Van Puyenbroeck and Rogge's (2017) 'indirect' multiplicative Benefit-of-the-Doubt (BoD) index number framework, in which the linear, data-driven BoD-model is used to estimate the importance of various sub-indicators within a geometric composite index (CI). We present an integrated framework that combines optimistic and pessimistic BoD-based weighting that enables to (1) establish the degree of unbalance in countries' policy portfolio mix, (2) identify multiple underlying factors to explain inter-temporal evolution, and (3) explain for differences in country policy performances under the different weighting schemes following a multiplicative Bortkiewicz decomposition. In doing so, we use alternative optimistic and pessimistic BoD-models as existing models suffer from potential drawbacks in the identification of a country's comparative strengths and weaknesses. We illustrate our results with social inclusion data for the EU-countries for the period 2008-2013.