About this publication
Official statistics play a critical role in supporting evidence-based decision making, particularly during severe crises characterized by high uncertainty and intensified user needs. In recent years, Lebanon has faced multiple overlapping shocks, notably a prolonged economic and financial crisis and the war during 2023–2024. These exceptional conditions posed major challenges to the continuity, credibility, and quality of official statistical production.
This paper presents the experience of the Central Administration of Statistics of Lebanon in maintaining quality standards across two key statistical domains: the compilation of National Accounts under extreme economic instability and the implementation of the Multiple Indicator Cluster Survey (MICS) under severe security constraints. The analysis is framed by the Code of Practice principles most relevant in crisis contexts relevance, accuracy, reliability, coherence, comparability, and transparency and examines how these principles were upheld despite disrupted data sources and operational limitations.
Under the combined effects of economic crisis and war-related disruptions, National Accounts compilation was challenged by high inflation, sharp currency depreciation, and the coexistence of multiple official and market exchange rates. Addressing these issues required consistent application of valuation methods, careful treatment of price and exchange rate effects, and explicit handling of depreciation. Delays and gaps in administrative data availability necessitated the use of well-documented estimation techniques and coherent assumptions. These were accompanied by transparent communication on methodologies, exchange rate choices, and data revisions to preserve user confidence in an environment of extreme uncertainty.
For the MICS survey, security constraints prevented the completion of fieldwork in three governorates. To ensure quality and transparency, results were published separately for each completed governorate, and national-level estimates were not produced to avoid bias from incomplete coverage. Sampling weights were recalculated for the completed strata, and all methodological limitations were clearly documented to maintain analytical validity and trust.
Overall, the paper highlights key challenges, solutions, and lessons learned, demonstrating that adherence to quality principles, transparency, and institutional resilience is essential for sustaining trust in official statistics during severe crises.