Atlas Data Downloads

API Access

Atlas users can query data directly through our GraphQL API. Documentation on access procedures and limitations is available on this GitHub page.

Growth Lab Trade Data Methodology

International trade flows are recorded twice: once by exporters and once by importers. In practice, these records often differ, are incomplete, or are reported using different product classifications. To address these issues, the Growth Lab developed a standardized, data-driven process to produce consistent and comparable bilateral trade estimates.

Our methodology combines two core steps: reconciling partner reports and harmonizing product classifications over time.

Reconciling Exporter and Importer Reports

We first adjust reported values to ensure comparability between exporter-reported (FOB) and importer-reported (CIF) data. We then assess the reliability of each country's reporting based on its consistency across trading partners. Using these reliability scores, we combine exporter and importer reports into a single estimated trade value that places greater weight on more reliable sources.

Harmonizing Product Classifications

Countries adopt new trade classification systems at different times, and product codes are regularly revised, split, or merged. To ensure comparability across years, we use data-driven conversion weights that translate trade values between classification vintages while preserving detailed product information. This approach avoids the loss of products and discontinuities that arise from simple one-to-one mappings.

By integrating reliability-weighted mirroring with weighted product conversion, we generate bilateral, product-level trade datasets that are consistent across countries and over time. These datasets form the core data underlying the Atlas and improve coverage, reduce reporting discrepancies, and support long-run analysis of global trade patterns.

A full description of this methodology is available in our peer-reviewed paper:

Bustos et al. (2026), Tackling Discrepancies in Trade Data: The Harvard Growth Lab International Trade Datasets, Scientific Data.

A visual description of the Growth Lab's trade data methodology is available on our companion website.

Data Sources

The Atlas uses data from the following international organizations as inputs into its comprehensive methodology:

• Goods Trade: United Nations Statistical Division (UN Comtrade)

• Services Trade: International Monetary Fund, Direction of Trade Statistics (DOTS)

• Economic Indicators: International Monetary Fund, World Economic Outlook (WEO)

• Price Adjustments: Federal Reserve Economic Data (FRED), Producer Price Index for Industrial Commodities

Atlas trade values are reported in U.S. dollars (USD). Constant-dollar values are adjusted using the FRED Producer Price Index for Industrial Commodities, with the base year set to the most recent Atlas data year. This allows users to interpret values in terms of current purchasing power.

For historical analysis (e.g., growth rates and long-run trends), constant-dollar values are recommended.

Data Classifications

Product Classifications

The Atlas provides trade data using two complementary classification systems:

Harmonized System (HS)

Covers approximately 5,000–5,600 products using 6-digit codes (exact counts vary by revision). Best suited for analyzing contemporary products and industries. HS classifications are available in the Atlas in three revisions:

• HS 1992: data available from 1995 onward

• HS 2012: data available from 2012 onward

• HS 2022: data available from 2022 onward

Standard International Trade Classification (SITC)

Provides a harmonized, long-run dataset spanning multiple classification systems from 1962 to the present. Covers approximately 700 products using up to 4-digit codes. Best suited for long-term historical analysis and trend comparison. SITC data in the Atlas is based upon SITC Rev. 2.

Services Trade

Services exports and imports are reported unilaterally, with partner countries grouped as "Services Partners." Coverage begins in 1980 and varies by country, with data available for approximately 50 to 75 percent of Atlas countries.

Country Coverage

While Atlas Explore includes data for all countries and territories covered by UN Comtrade, Atlas Country Profiles and Rankings are limited to countries that meet minimum coverage and quality standards:

• Population of at least 1 million

• Average annual trade volume of at least $1 billion

• Verified GDP and export data availability

• Consistent and reliable trade reporting history

The Atlas aims to reflects countries as they exist today, following UN recognition. Map boundaries come from United Nations geospatial data, chosen for its stability over time and its alignment with our trade data sources. When country names, borders, or political status change, the Atlas may take time to reflect those updates as we revise the underlying data and visualisations.

Data Update Cycle

Atlas data updates follow international trade reporting cycles and require substantial processing to ensure quality and consistency.

Annual Updates

Approximately 95 percent of Atlas data is updated once per year, typically between April and June. This reflects country reporting timelines to UN Comtrade, which generally require 12 to 18 months to reach sufficient coverage. For example, most 2024 trade data becomes available in the Atlas between April and June 2026.

Annual releases may incorporate late or corrected country reporting which can lead to small revisions in historical data. This process improves long-run accuracy and consistency.

Interim Updates

Throughout the year, we also release interim updates that incorporate newly submitted country data, reflect improvements in data processing methods, and address minor corrections when needed.

For real-time information on country-level data availability, see the UN Comtrade Data Availability Dashboard.

Usage and Citation

The Atlas of Economic Complexity is a freely available public resource. Its data and tools are widely used in research, policy analysis, and applied projects. We encourage reuse and adaptation, provided that users cite the Atlas and underlying data sources appropriately.

Citation Guidelines

When incorporating Atlas content in your work, please use these citation formats:

For the Atlas Platform:

Growth Lab at Harvard University. "The Atlas of Economic Complexity." Web application. Harvard Kennedy School, 2024. https://atlas.hks.harvard.edu

For Downloaded Datasets:

Each downloadable data file includes specific citation requirements.

For the Original Atlas Book:

Hausmann, R., Hidalgo, C., Bustos, S., Coscia, M., Chung, S., Jimenez, J., Simoes, A., Yildirim, M. (2013). The Atlas of Economic Complexity. Cambridge, MA: MIT Press.

Support

We provide several resources to help users understand and work with Atlas data:

• Glossary: Definitions of key terms, indicators, and classifications

• FAQ: Answers to common questions about data sources, methods, and interpretation

• GitHub Repositories: Technical documentation and data access tools

For technical questions, research use cases, or data access requests, contact us at growthlabtools@hks.harvard.edu.