FAO FAOSTAT API and bulk downloads

Use in your agent

Load a FAOSTAT consumer-price bulk file into pandas and Parquet. Checked against the provider's documentation on Nov 1, 2026.

Access

• No account or key is needed for the checked bulk catalog and ZIP file.
• The API developer portal documents programmatic JSON and CSV access; the API host was intermittent during this check.

Request

curl -O 'https://bulks-faostat.fao.org/production/ConsumerPriceIndices_E_All_Data_(Normalized).zip'

Format

ZIP containing normalized CSV observations plus area, item, element and flag lookup files. Observation columns include area, item, element, month, year, unit, value, flag and note.

Load into pandas and Parquet

Needs Python 3 with requests, pandas and pyarrow.

"""FAOSTAT normalized consumer-price bulk file -> pandas -> Parquet."""
import io
import zipfile
import requests
import pandas as pd

URL = "https://bulks-faostat.fao.org/production/ConsumerPriceIndices_E_All_Data_(Normalized).zip"
HEADERS = {"User-Agent": "Mozilla/5.0 SignalLabCheck/1.0"}
r = requests.get(URL, headers=HEADERS, timeout=120)
r.raise_for_status()
with zipfile.ZipFile(io.BytesIO(r.content)) as zf:
    name = "ConsumerPriceIndices_E_All_Data_(Normalized).csv"
    df = pd.read_csv(zf.open(name), low_memory=False)
df.to_parquet("faostat_consumer_prices.parquet", index=False)
print(len(df), "rows", df["Area"].nunique(), "areas", int(df["Year"].min()), "to", int(df["Year"].max()))

Run by us on 2026-11-01 (python 3.13.5 | pandas 3.1.0 | pyarrow 26.0.0 | requests 2.34.2). Result: 266292 rows 247 areas 2000 to 2026

Limits

• Rate limit: Not stated.
• The checked normalized consumer-price ZIP is about 2.1 MB compressed and expands to about 41.3 MB across five CSV files.
• Bulk files contain complete domains; select columns and rows after download.

Quirks

• Domain release dates differ; use the bulk catalog DateUpdate field rather than assuming one FAOSTAT-wide refresh date.
• Keep Flag and Note because they identify estimates, imputations and base-year details.
• Area Code (M49) contains a leading apostrophe in the CSV and is safest to retain as text.

Terms and attribution

• FAOSTAT data are CC BY 4.0 unless a dataset's metadata or page says otherwise.
• Use FAO's required database citation and check metadata for third-party exceptions.

Sources

• https://www.fao.org/statistics/highlights-archive/highlights-detail/faostat-launches-a-new-api-developer-portal-to-make-data-access-easier/en: “Access all FAOSTAT domains through a single, consistent interface”
• https://www.fao.org/statistics/highlights-archive/highlights-detail/faostat-launches-a-new-api-developer-portal-to-make-data-access-easier/en: “Filter data by area, item, element, year and more dimensions”
• https://www.fao.org/contact-us/terms/db-terms-of-use/en: “all datasets disseminated through FAO corporate statistical databases (see examples in Annex 1) are licensed under the Creative Commons Attribution-4.0 International licence”