Hosting Capacity and Economic Viability of Grid-Connected Photovoltaic Systems in Mozambique: A Finance-First Framework Analysis Based on 1,320 Observations (2015-2024)

When the discount rate, not solar irradiance, is the dominant predictor of the levelized cost of energy in one of the world's sunniest countries, energy policy needs recalibrating. Across 110 grid-connected photovoltaic projects in Mozambique (1,320 observations, 2015-2024), OLS regressions, Logit models, and fixed-effects panel estimators consistently identify financial structure, not technical performance, as the binding constraint on PV viability. This article formalizes that empirical regularity into the Finance-First PV Integration Framework (F²PIF): a three-stage sequential architecture in which financial risk mitigation is established before not alongside technical site assessment. Key findings are: (1) the discount rate is the primary LCOE predictor (β =-0.003, p < 0.05) while GHI, installed capacity, and technology type are statistically indistinguishable from zero; (2) the Performance Ratio is the sole significant technical predictor of financial viability (Logit OR = 31.4, p < 0.10), identifying preventive maintenance as the highest-return operational intervention; (3) rural network hosting capacity reaches 15% of peak load under current conditions but approximately doubles with 2-hour BESS integration; and (4) regulatory gaps not grid physics represent the binding scalability constraint. These findings challenge the techno-centric paradigm in energy access planning and offer a replicable framework for Mission 300 and SDG 7 implementation across sub-Saharan Africa.