Skip to content

Aggregate Steel Demand | Bottom-Up Assessment Framework

Bottom-Up Steel Demand Calculation Methodology

Author: Dr Andrzej M Kotas | ORCID: 0009-0009-5497-5384

Methodology Context: This framework has been developed and refined through 25+ years of steel industry consulting to governments, development banks, and steel producers across Africa, Eastern Europe, and global markets. The approach supports strategic advisory work requiring accurate demand assessment for policy development, capacity planning, and investment analysis.

Key Terms

ADC
Apparent Domestic Consumption — steel demand estimated as Production + Imports - Exports
GAC
Gross Apparent Consumption — ADC before adjusting for downstream domestic processing
Net Consumption
GAC minus tonnage consumed by domestic processors — avoids double-counting across value chain stages
HS Codes
Harmonised System product classification codes used to extract steel trade data by product category
Mirror Data
Trading partner import/export statistics used to fill gaps where direct country data is unavailable
WSA
World Steel Association — primary source for global steel production statistics

Overview

This methodology provides a systematic framework for calculating regional and country-level steel demand using a bottom-up apparent domestic consumption (ADC) approach. Unlike top-down GDP multiplier methods, this granular analysis examines steel consumption at the product level, incorporates comprehensive trade data triangulation, and applies multiple integrity cross-checks to ensure accuracy.

The framework is designed for:

Core Calculation Framework: Apparent Domestic Consumption (ADC)

Apparent Domestic Consumption (ADC) Calculation Framework showing production plus imports minus exports equals ADC

Finished Steel ADC Calculation Framework with Data Triangulation

The fundamental equation for apparent domestic consumption is:

Frequently Asked Questions

Apparent domestic consumption (ADC) is calculated as: Production + Imports - Exports. It represents the best available estimate of actual steel demand within a specific geographic market, capturing total steel availability for domestic use after accounting for cross-border trade flows. ADC differs from actual end-use consumption because it does not capture inventory changes or stockpiling behaviour.
Steel demand is calculated bottom-up by analysing consumption at the product level for each steel category (rebar, HRC, CRC, plate, sections, etc.) using the ADC formula: Production + Imports - Exports. Gross consumption figures are then adjusted for downstream processing to avoid double-counting, and the sum of net consumption across all products gives total steel demand. This approach is more granular and accurate than top-down GDP multiplier methods.
Gross apparent consumption is the initial ADC calculation (Production + Imports - Exports) for each product category. Net consumption adjusts this figure by subtracting tonnage consumed by downstream domestic processors to avoid double-counting. For example, hot-rolled coil consumed by domestic cold-rolling mills is subtracted from HRC net consumption and counted instead in CRC consumption figures. The sum of all net consumption figures gives total steel demand without duplication.
The primary data sources are: World Steel Association (WSA) production statistics, which cover approximately 85% of world steel output; UN Comtrade database for bilateral trade flows using Harmonised System (HS) codes; International Steel Statistics Bureau (ISSB) for trade data validation; steel plant capacity databases to cross-check production figures; and mirror image trade data from partner countries where direct statistics are unavailable.
Harmonised System (HS) codes are internationally standardised product classification codes used in customs and trade statistics. In steel demand analysis, specific HS codes are assigned to each steel product category (plate, HRC, CRC, galvanised sheet, rebar, rod, sections, seamless tube, welded tube, etc.) to extract import and export data from databases such as UN Comtrade and ISSB. This ensures consistent product classification across countries and years.
Mirror image trade data is used when a country's import or export statistics are missing or unreliable. If Country A's export data to Country B is unavailable, Country B's import data from Country A is used as a proxy estimate. Every export from Country A to Country B should appear as a corresponding import in Country B's statistics. Significant discrepancies between mirror image figures and direct reports indicate data quality issues requiring investigation.
Eight independent integrity checks are applied: (1) production vs. capacity validation (production rarely exceeds 90% of installed capacity); (2) aggregate finished steel consumption cross-check; (3) comparison against World Steel Association benchmark figures; (4) per capita consumption benchmarking against comparable economies; (5) product mix ratio analysis (flat vs. long vs. tubular products); (6) negative consumption detection; (7) year-over-year volatility analysis; and (8) crude steel gross-up verification working backward from finished products to crude steel output.
Key limitations of ADC methodology include: statistical reporting lags of 1-6 months; inability to capture inventory changes (ADC measures availability not actual consumption); distortion from re-exports and trans-shipment hubs such as Singapore or Dubai; absence of reliable statistics for some countries requiring indirect estimation methods; changes in HS code definitions affecting multi-year trend analysis; and periodic data revisions that may require retroactive adjustment of historical demand assessments.

Key Definitional Concepts

  • Production: Domestic steel output for the analysis period, typically sourced from World Steel Association data or national statistics agencies. May be validated against capacity data adjusted for utilisation rates.
  • Imports: Steel products entering the domestic market from international sources, captured through customs trade data using HS codes specific to steel products.
  • Exports: Domestically-produced steel shipped to international markets, similarly tracked through trade statistics.
  • Apparent Consumption vs. Actual Consumption: ADC provides demand estimates at the point of first domestic availability. Actual end-use consumption may differ due to inventory changes, stockpiling behaviour, and distribution system dynamics. ADC serves as the most reliable proxy for demand when direct consumption data is unavailable.

Data Sources & Collection Methodology

Primary Data Sources

  • World Steel Association Production Statistics: Official steel production data by country and product category. WSA data represents the authoritative global standard for production statistics, compiled from member submissions covering approximately 85% of world steel output.
  • United Nations Comtrade Database: Comprehensive international trade statistics using Harmonised System (HS) codes for steel product classification. Provides bilateral trade flow data between all reporting countries.
  • International Steel Statistics Bureau (ISSB): Alternative source for trade statistics with detailed HS code breakdowns for steel products. Particularly valuable for validation and gap-filling where UN Comtrade data is incomplete.
  • Capacity & Utilisation Data: Steel plant capacity information from James King steel capacity database, Metal Bulletin's Iron & Steel Works of the World, Metal Expert database, and direct company sources. Used to validate production figures through capacity utilisation calculations (production rarely exceeds 90% of nominal capacity).
  • Mirror Image Trade Data: When import or export data is missing for a specific country, corresponding export or import data from trading partner countries provides alternative estimates. For example, if Country A's export data to Country B is unavailable, Country B's import data from Country A serves as a proxy.

HS Code Framework for Steel Products

Steel trade data is collected using internationally standardised HS codes covering major product categories:

Plate
Hot-Rolled Coil (HRC)
Cold-Rolled Coil (CRC)
Hot-Dip Galvanised (HDG)
Pre-Painted Galvanised (PPGI)
Rail
Sections
Rebar
Rod
Other Bar
Seamless Tube
Welded Tube

HS codes and detailed steel product classifications are available from ISSB's international steel trade statistics portal.

Product-Level Analysis: Gross vs. Net Consumption

A critical refinement distinguishes between gross apparent consumption and net consumption to avoid double-counting in vertically integrated steel value chains:

Gross Apparent Consumption (GAC)

Initial ADC calculation for each product category using the standard formula: Production + Imports - Exports. This represents total product availability before accounting for downstream processing.

Net Consumption Adjustment

For products undergoing further domestic processing, gross consumption must be adjusted to reflect actual end-use consumption:

  • Example 1 - Hot-Rolled Coil: If a country produces 5Mt of HRC domestically, imports 1Mt, and exports 0.5Mt, gross ADC is 5.5Mt. However, if 3Mt of that HRC is consumed by domestic cold-rolling mills to produce CRC, net HRC consumption (for direct end-use) is only 2.5Mt. The 3Mt is captured in downstream CRC consumption.
  • Example 2 - Cold-Rolled Coil: Similarly, CRC consumed by domestic galvanising lines must be subtracted to calculate net CRC consumption, with that tonnage captured in HDG/PPGI consumption figures.

This adjustment prevents counting the same steel tonnage multiple times across different processing stages. The sum of all net consumption figures across finished products accurately represents total steel demand without duplication.

Data Triangulation & Quality Assurance

Robust demand assessment requires systematic cross-validation using multiple independent data sources and analytical relationships:

Production Validation

✓ Capacity Utilisation Cross-Check

Reported production volumes are validated against installed capacity data. Steel mills typically operate at 70-90% of nameplate capacity, rarely exceeding 90% except during exceptional market conditions. Production figures significantly above capacity thresholds trigger investigation for data accuracy or unreported capacity additions.

Trade Data Validation

✓ Mirror Image Verification

Every export movement from Country A to Country B should appear as a corresponding import in Country B's statistics. When direct import/export data is missing, mirror image data from trading partners provides alternative estimates. Significant discrepancies between mirror image figures and direct reports indicate potential data quality issues requiring further investigation.

✓ Bilateral Trade Flow Consistency

Cross-checking reported exports from Source Country to multiple destinations against those destinations' import data reveals systematic over-reporting or under-reporting patterns. Persistent discrepancies may indicate trans-shipment hubs, customs valuation differences, or statistical reporting gaps.

Comprehensive Integrity Cross-Checks

Multiple analytical cross-checks ensure internal consistency and identify potential data errors:

1. Production vs. Capacity Validation Production volumes exceeding 90% of nominal capacity warrant verification, as such high utilisation is uncommon except during peak demand periods.
2. Aggregate Finished Steel Consumption Sum of net consumption estimates across all finished steel products provides total finished steel demand. This aggregate figure enables comparison against top-level demand assessments.
3. World Steel Association Benchmark Comparison Total consumption calculated through bottom-up product analysis is compared against WSA's aggregate apparent consumption figures for the same country/region. Material deviations require investigation.
4. Per Capita Consumption Benchmarking Demand intensity measured in kg per capita should align with similar economies at comparable development stages. Significant outliers suggest data inconsistencies or unique structural factors requiring explanation.
5. Product Mix Ratio Analysis Proportions of flat products vs. long products vs. tubular products tend toward consistent patterns within similar economic structures. Unusual product mix distributions may indicate incomplete data coverage or specialised market characteristics.
6. Negative Consumption Detection Negative apparent consumption figures (exports exceeding production plus imports) indicate data errors in one or more components. Such anomalies require immediate investigation and correction.
7. Year-Over-Year Volatility Analysis Large unexplained swings in individual product consumption from one year to another may highlight calculation errors, statistical revisions, or genuine market disruptions requiring documentation.
8. Crude Steel Production Gross-Up Verification Working backward from finished products through intermediate products (HRC ? slab) to crude steel provides an independent production estimate. This reverse calculation should reconcile with reported crude steel output within yield loss tolerances.

Methodological Limitations & Data Quality Considerations

⚠ Known Limitations

  • Statistical Reporting Lags: Official production and trade statistics typically lag actual activity by 1-3 months for monthly data, up to 6 months for annual data. Real-time demand assessment requires provisional estimates subject to later revision.
  • Inventory Changes Not Captured: ADC measures availability rather than actual consumption. Significant inventory building or drawdown periods can cause ADC to diverge from true end-use demand. This effect is particularly pronounced during economic transitions or trade policy changes.
  • Re-Export and Trans-Shipment Complexity: Products imported for immediate re-export inflate apparent consumption figures. Major trans-shipment hubs (Singapore, Dubai, etc.) require careful analysis to distinguish transit trade from genuine domestic consumption.
  • Non-Reporting Countries: Some jurisdictions do not publish reliable production or trade statistics. For these markets, demand estimates rely on indirect methods including construction activity proxies, automotive production data, or regional trade partner analysis.
  • Product Classification Variations: HS code definitions occasionally change, and countries may implement codes with slight variations. Multi-year trend analysis requires careful reconciliation of product category definitions.
  • Data Revision Cycles: Both production and trade data undergo periodic revisions as more complete information becomes available. Historical demand assessments may require retroactive adjustment when major revisions are published.

Applications of Demand Methodology

This bottom-up ADC framework serves multiple analytical purposes across steel industry stakeholders:

Policy Development

Governments use accurate demand assessment to evaluate domestic steel industry development needs, import dependency levels, and strategic material security considerations. Demand forecasts inform infrastructure planning and industrial policy design.

Capacity Planning

Steel producers require granular product-level demand analysis to guide capacity expansion decisions, product mix optimisation, and facility modernisation investments. Understanding regional demand patterns enables strategic facility location decisions.

Investment Due Diligence

Development banks and private equity investors conducting steel sector due diligence need rigorous demand assessment to evaluate market viability, competitive intensity, and demand-supply balance. This methodology provides the analytical foundation for investment thesis development. Where demand forecasting has been weak or optimistic, the consequences can be severe — the Steel Industry Investment Disasters podcast episode examines four major cases where flawed market assumptions contributed to catastrophic project outcomes.

Trade Policy Analysis

Assessment of import penetration rates, export dependency, and bilateral trade patterns supports trade policy evaluation and negotiation. Understanding consumption patterns by product helps identify sectors vulnerable to import competition or export market opportunities.

Market Research

Commercial market research organisations apply this framework to produce steel demand forecasts for subscription services, consulting projects, and industry outlook reports. Product-level granularity enables sector-specific demand modelling.

Academic Research

Researchers studying steel industry economics, trade patterns, industrial development, or commodity market dynamics utilise transparent demand calculation methodologies to ensure replicable analysis and peer review.

Methodology Evolution & Updates

This analytical framework has been continuously refined since 2001 through practical application across diverse markets and economic conditions. The methodology incorporates lessons learned from:

  • Steel demand assessment projects for governments in Africa and Eastern Europe
  • Due diligence assignments for development banks evaluating steel sector investments
  • Market entry studies for steel producers expanding into new geographic markets
  • Expert witness work requiring defensible demand quantification for legal proceedings.

The framework adapts to evolving data availability, statistical reporting standards, and analytical requirements while maintaining core principles of transparency, triangulation, and rigorous verification.

Methodology Quality Standards

Data Source Transparency: All sources documented & cited
Calculation Replicability: Formula disclosed, results reproducible
Limitation Acknowledgment: Known constraints explicitly stated
Cross-Validation: 8 independent integrity checks applied
Professional Application: Used in institutional consulting since 2001

Methodology Disclosure & Professional Standards

This methodology documentation reflects SteelOnTheNet's commitment to analytical transparency and professional standards. The framework has been developed through Dr Andrzej M Kotas's 25+ years of steel industry consulting experience across multiple continents and regulatory environments.

The approach aligns with institutional research standards employed by multilateral development banks, government statistical agencies, and professional consulting firms. By publishing this methodology, we enable clients, researchers, and industry stakeholders to understand the analytical foundation supporting our strategic intelligence services.

Related Methodologies

Questions about this methodology? For detailed discussions about application to specific markets, data source selection, or custom analytical requirements, please contact our team.