How BGI Works
Framework
In 2008, the President of France commissioned Joseph Stiglitz, Amartya Sen, and Jean-Paul Fitoussi — two Nobel laureates and a leading economist — to answer a simple question: if GDP doesn’t measure wellbeing, what should? Their report identified 8 dimensions of life that matter. It did not, however, build an index. BGI is one implementation of the Stiglitz-Sen-Fitoussi framework. The 8 dimensions come directly from the Commission’s work. What BGI adds is the indicator selection (27 indicators mapped to pillars), the scoring methodology (weighted geometric mean with floor thresholds), confidence intervals, and interactive weight adjustment so users can see how different value judgments shift the rankings.
The 8 Dimensions
Scoring
BGI uses a weighted geometric mean to combine pillar scores. Unlike an arithmetic mean, a geometric mean penalizes imbalance — a country can’t compensate for terrible environmental performance with high GDP. Near-zero in any dimension drags the entire score down.
Floor Thresholds
Some failures are non-negotiable. A pillar score below 25 triggers a Critical flag. Below 40, a Warning. These flags appear regardless of the overall BGI score — you can’t average away a crisis.
Confidence Intervals
Every BGI ranking includes Monte Carlo confidence intervals. We simulate 10,000 weight configurations around the default, producing a distribution of possible scores. This tells you whether a country’s rank is robust or fragile.
Open Source
The entire BGI scoring engine is open source. Every score can be reproduced from raw data using public code. We believe transparency is the only foundation for trust in composite indices.
BGI is an independent research project and not affiliated with the OECD, World Bank, or any government body. Methodology subject to revision as data sources improve.