Input–Output (I‑O) Analysis

Historical Origins

Input–Output analysis was formalised by Wassily Leontief in the 1930s as a response to the need for systematic measurement of industrial interdependence in modern economies. Leontief’s tables translated production processes into a matrix of sectoral transactions, enabling analysts to trace how output in one industry requires inputs from others. Initially applied to national planning and wartime production logistics, I‑O methods spread into development economics and regional planning as statistical systems matured. Over time the technique was adapted to include environmental extensions, satellite accounts for natural resources, and regional disaggregation—each adaptation increasing its relevance for resource‑intensive sectors.

What I‑O Analysis Does

I‑O analysis quantifies direct, indirect, and induced economic effects by mapping the flow of goods and services across sectors. It produces multipliers for output, employment, and value added, and identifies critical supplier linkages and import leakages. For extractive projects, I‑O provides a first‑order estimate of the economy‑wide footprint: how mine output translates into demand for construction, transport, energy, and local services, and how wages and procurement circulate through households and firms.

Why this is important for the extractive industries

• Value‑capture diagnostics: I‑O reveals where value is created and where it leaks abroad, informing local‑content and supplier‑development strategies.

• Regional planning: it identifies sectors that will expand or contract with new projects, guiding infrastructure and workforce planning.

• Short‑term fiscal estimates: governments use I‑O multipliers to estimate immediate revenue and employment impacts for budgeting and compensation design.

• Baseline for dynamic models: I‑O tables often serve as the calibration backbone for SAMs and CGE models used in longer‑term analysis.

  
  

Main methodological challenges and considerations

• Static structure: technical coefficients are fixed, so I‑O cannot capture price responses, substitution, or technological change; it overstates long‑run impacts if used alone.

• Data quality and vintage: national and regional I‑O tables are frequently outdated and lack mine‑level granularity; imported input shares must be carefully adjusted.

• Aggregation bias: sectoral aggregation can mask commodity‑specific supply chains and local heterogeneity.

• No behavioural feedbacks: labour mobility, wage inflation, and exchange‑rate effects are outside the model’s scope.

• Mitigation strategies: use I‑O for short‑to‑medium‑term linkage diagnostics; combine with SAMs for distributional detail and with CGE or econometric models to capture price and behavioural dynamics; apply sensitivity analysis to multipliers and explicitly adjust for import content.

  
  

Practical guidance

Use I‑O analysis as a diagnostic first step: produce transparent multipliers, identify priority supplier sectors, and quantify immediate regional impacts. Do not rely on I‑O alone for long‑term policy design; instead, integrate its outputs into dynamic models and participatory assessments to inform robust, implementable extractive‑sector strategies.