Configurational Drivers of Fiscal Concentration: An fsQCA Analysis Using the APG Fiscal Concentration Index

Fiscal monitoring typically evaluates debt, deficits, expenditure, guarantees, and revenue capacity as separate indicators. Such measures are indispensable, but they do not by themselves explain why an entire fiscal profile becomes structurally concentrated. This study extends the Arithmetic Power Geometry Fiscal Concentration Index (AFCI) from measurement to configurational explanation. AFCI combines the maximum squared-normalized fiscal weight with normalized entropy loss and is calculated for a CAG-derived panel of 28 Indian states from 2013–14 to 2022–23. The paper first establishes nine theorems concerning AFCI: probability-simplex validity, scale invariance, permutation invariance, sharp bounds, continuity, majorization monotonicity, the equalizing-transfer principle, complete-concentration convergence, and generalized-index validity. It then applies fuzzy-set Qualitative Comparative Analysis to 251 temporally ordered state-year observations, using prior-year debt burden, fiscal deficit burden, committed expenditure, guarantees, capital orientation, and revenue adequacy as conditions for current-year high or low AFCI. The conservative high-concentration solution attains consistency of .925 and coverage of .783; every principal pathway contains high debt burden, but debt combines with different forms of deficit pressure, expenditure rigidity, guarantee exposure, weak revenue adequacy, or low capital orientation. The independently estimated low-AFCI solution has consistency of .943 and coverage of .790, demonstrating set-theoretic configurational asymmetry rather than simple inversion. Robustness analyses across calibration anchors and consistency thresholds preserve the central role of debt burden. The contribution is threefold: a formal axiomatic validation of AFCI, a temporally ordered configurational account of fiscal concentration, and an Apache-2.0 reproducibility package that regenerates the data, calibrations, truth tables, solutions, tables, and figures.