Skip to main content
v2026.11,986 entries · CC-BY 4.0

IMF Data: Datasets, WEO Database, API and Reuse Terms

What IMF data is, which datasets it covers, how to access the WEO and API, the October 2025 code change, and the attribution rules for reuse.

Written and maintained by CASRAI Editorial Board

Published

On this page

IMF data is the published statistical data that the International Monetary Fund produces or curates, released through the IMF Data Portal (data.imf.org) and the IMF’s Data pages. It covers macroeconomic and financial series such as the World Economic Outlook (WEO) projections, national accounts, balance of payments, consumer prices and financial access, and it can be browsed, downloaded in bulk or queried through an API. Researchers can download, copy, adapt and republish it, provided the IMF is credited as the source and any transformation is stated.

What is IMF data?

The IMF’s copyright and terms page defines “Data” as published statistical data produced or curated by the IMF. That wording matters for research use: some series are compiled by the IMF itself (the WEO forecasts are the best-known example), while others are statistics that member countries report and the IMF curates and publishes. In either case the data you download from an IMF site falls under the same Special Terms for Statistical Data, covered in the reuse section below.

The IMF describes its Data Portal as the system of record for publicly released IMF datasets, replacing earlier legacy dissemination systems. For a researcher this means the portal is the place to look for the current official version of a series, and that older bookmarks to legacy download pages may no longer point at current data. The WEO is the clearest case: its database moved from a page under imf.org/en/publications/weo/weo-database to a dataset page on the Data Portal (IMF.RES:WEO) between April 2025 and October 2025.

Which datasets does the IMF publish?

The IMF’s Data page lists six headline datasets as quick links, and several tools built on top of them. The data table below this guide sets them out with the detail the IMF gives for each. In summary:

  • World Economic Outlook databases: the IMF’s macroeconomic projections, published with the April and October WEO reports.
  • National Economic Accounts, Balance of Payments and International Investment Position, International Financial Statistics, Consumer Prices and the Financial Access Survey: the other five headline datasets.
  • IMF DataMapper: a visualisation and download tool covering 13 key datasets.
  • Macroeconomic Climate Indicators Dashboard: climate-related indicators in six categories (greenhouse gas emissions, mitigation, adaptation, transition to a low-carbon economy, climate finance, and climate and weather).
  • PortWatch: monitoring of maritime trade disruption.
  • Coordinated Direct Investment Survey (CDIS): direct investment statistics.

The IMF’s finance data (members’ financial data, SDR valuation and interest rate, among others) is published separately from the economic statistics. If your question concerns trade flows between named countries rather than IMF macro aggregates, see the guide to UN Comtrade, which is a different producer and data model.

How do you access IMF data?

The IMF’s WEO transition guideline (current as of 9 October 2026) describes three access modes for the WEO dataset:

  • Interactive selection. Choose countries, indicators and years and use “View Data” to see a table, then download it.
  • Bulk download. The full WEO database can be downloaded for the latest vintage, which is the April or October release.
  • API. The IMF states that its data are available through SDMX 2.1 and SDMX 3.0 APIs, with guidance for MATLAB, Python, R and Stata users. The IMF’s help desk handles API questions. The transition guideline also mentions an Excel add-in as forthcoming.

Pick the mode by how often you will repeat the pull. A one-off chart can come from the interactive view; a script that is re-run every WEO cycle should use the API or the bulk file so the retrieval is reproducible. For a wider look at where datasets like these sit among open repositories, see how to choose an open data repository.

How do WEO series codes work after October 2025?

This is the change most likely to break an older script. From the October 2025 WEO onward, the WEO dataset uses the three-letter ISO country code as the country identifier, where it previously used the three-digit IFS country code. Series codes (SERIES_CODE) are built with the ISO code. The IMF’s own example is USA.NGDP_RPCH.A, which is real GDP growth for the United States at annual frequency.

Part of the code In the example What it identifies
Country USA Three-letter ISO country code (replaces the three-digit IFS code from the October 2025 WEO)
Indicator NGDP_RPCH The indicator; in this example, real GDP growth
Frequency A Annual, in the IMF’s example

The new dataset also carries expanded metadata fields, including COUNTRY.ID, INDICATOR.ID, UNIT, SCALE, LATEST_ACTUAL_ANNUAL_DATA, METHODOLOGY, BASE_YEAR and BASIS_OF_PROJECTIONS. Reading UNIT and SCALE before you plot or model a series prevents the most common error with macro data, which is mixing percentages, index numbers and currency amounts in different scales.

Worked example: reading one series code

Suppose you want annual US real GDP growth from the WEO. Using the structure above, the series code is USA.NGDP_RPCH.A: USA for the country, NGDP_RPCH for the indicator and A for annual. If an older script keyed the United States by its three-digit IFS code, it will no longer match the October 2025 and later releases, and the lookup must be rewritten to use the ISO code. Then check LATEST_ACTUAL_ANNUAL_DATA for that series: values after the last actual year are projections, and the BASIS_OF_PROJECTIONS field describes how they were formed. A growth rate that looks like an observation but sits beyond that year is a forecast and should be labelled as one.

What is a WEO vintage and why does it matter?

The WEO dataset (IMF.RES:WEO) is updated only at the April and October WEO publications, and a bulk download gives you the latest vintage. Each release can revise earlier projections and historical figures, so two researchers who downloaded “WEO data” six months apart may hold different numbers for the same country and year. Record the release (for example “October 2025 WEO”) alongside the download date, keep the raw file, and note it in your methods section. If you need to reproduce a published analysis, you must be able to say which vintage it used.

Can you reuse and cite IMF data?

Yes, within the IMF’s Special Terms for Statistical Data. Users may download, extract, copy, create derivative works from, publish, distribute and use Data obtained from IMF sites. The conditions that matter for researchers are:

  • Attribution. Distributed or reproduced data must appear accurately with attribution to the IMF as the source.
  • Transformations. If you materially transform the data (rebasing, merging with other sources, seasonal adjustment, your own aggregates), the transformation must be stated explicitly alongside the source citation.
  • Selling data. If you offer the data as a standalone product, you must tell purchasers that it is available free from the IMF.
  • Commercial reuse. For any potential commercial reuse, the IMF asks you to request permission from [email protected] (write to that address; do not rely on this page as a licence).
  • Scope. Free access does not give you a right to confidential or unpublished data.

A practical citation records the IMF as author, the dataset name (for example World Economic Outlook database), the release or vintage, the URL of the dataset page and the date you accessed it. Follow your journal’s or style guide’s template for the exact punctuation; the IMF content of the citation is what the terms require.

Common mistakes with IMF data

  • Treating projections as observations. WEO series mix actuals and projections; check LATEST_ACTUAL_ANNUAL_DATA.
  • Ignoring the vintage. Cite the April or October release, not just “the WEO”.
  • Using stale identifiers. Scripts that depend on three-digit IFS codes for WEO need updating to ISO codes for October 2025 and later data.
  • Following legacy links. Legacy download pages have been replaced; the Data Portal is the system of record.
  • Dropping the transformation note. If you recompute, rebase or merge series, say so, as the terms require.
  • Mixing units. Read UNIT and SCALE for every indicator before comparing countries.

IMF data and World Bank data

Both institutions publish country-level macroeconomic series, but they are different producers with different catalogues, identifiers and release cycles. IMF data is what the IMF itself produces or curates and publishes through its Data Portal; for the World Bank’s catalogue and access routes see the guide to World Bank data. When you combine the two in one analysis, cite each source separately, state which vintage of each you used, and note any harmonisation you applied. If you plan to model the combined series, the introduction to econometrics covers the methods side.

Frequently asked questions

Is IMF data free?

Yes. The IMF’s terms allow users to download, copy, adapt, publish and distribute published statistical data from IMF sites, provided they attribute the IMF and state any material transformation. Confidential or unpublished data is not covered.

How often is the WEO database updated?

The WEO dataset on the IMF Data Portal (IMF.RES:WEO) is updated only at the April and October WEO publications.

Does the IMF have an API?

Yes. IMF data is available through SDMX 2.1 and SDMX 3.0 APIs, and the IMF provides guidance for MATLAB, Python, R and Stata users. Questions go to the IMF Data Help Center.

Can I use IMF data commercially?

For any potential commercial reuse the IMF asks you to request permission from [email protected]. If you sell data as a standalone product you must tell purchasers that it is free from the IMF.

What changed in the WEO database in October 2025?

The three-letter ISO country code replaced the three-digit IFS code as the country identifier, series codes such as USA.NGDP_RPCH.A are built with the ISO code, and the dataset gained expanded metadata fields.

Sources

Data: IMF Data page: headline datasets and tools with the detail the IMF states for each

Compilation of the datasets and tools listed on the IMF Data page and the WEO transition guideline, with the detail the IMF gives for each entry.

IMF Data page: headline datasets and tools with the detail the IMF states for each
Dataset or toolKindDetail stated by the IMF
World Economic Outlook DatabasesDatasetHeadline dataset on the IMF Data page; WEO dataset IMF.RES:WEO is updated only at the April and October WEO publications
National Economic AccountsDatasetHeadline dataset listed as a quick link on the IMF Data page
Balance of Payments and International Investment PositionDatasetHeadline dataset listed as a quick link on the IMF Data page
International Financial StatisticsDatasetHeadline dataset listed as a quick link on the IMF Data page
Consumer PricesDatasetHeadline dataset listed as a quick link on the IMF Data page
Financial Access SurveyDatasetHeadline dataset listed as a quick link on the IMF Data page
IMF DataMapperVisualisation and download toolCovers 13 key datasets
Macroeconomic Climate Indicators DashboardDashboardSix categories: GHG emissions, mitigation, adaptation, transition to low-carbon economy, climate finance, climate and weather
PortWatchMonitoring toolMaritime trade disruption monitoring
Coordinated Direct Investment Survey (CDIS)Survey datasetListed among the IMF Data page tools and datasets

Sources

Compiled by CASRAI from the sources above. Free to reuse under CC BY 4.0 with attribution.

Follow CASRAI

Research-administration guidance, standards updates and independent tool reviews.

Ask CASRAI · free to try

Ask about IMF Data: Datasets, WEO Database, API and Reuse Terms

2 free questions a day, no account.

An AI assistant for research: funder rules, compliance, methods and publishing. It links the sources behind its answers and labels answers that come from a web search.

Answers draw on CASRAI's guides and dictionary plus the federal and funder documents we index: Federal Register, Grants.gov, Regulations.gov and UKRI.

Works on this site and inside Claude, Cursor and the AI tools you already use.

Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays open to read, with no account and no card.

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →