OICA World Motor Vehicle Production Statistics

Every country that builds cars, one table per year, since 1999. OICA - the International Organization of Motor Vehicle Manufacturers - collects production unit counts from its member associations and national correspondents and publishes them as country-and-region breakdowns split into passenger cars and commercial vehicles (light commercial vehicles, heavy trucks, coaches and buses). Each release carries a heading like 'World Motor Vehicle - Production - 2024' or 'Passenger Cars - Production - 2006', closes with a World Total row, and includes partial-year quarterly releases such as 2026 Q1 alongside the annual series.

Two things make this series usable rather than just quotable. First, the definitions are written down: passenger cars are road motor vehicles other than motorcycles intended to carry passengers, seating no more than nine people including the driver - which sweeps in taxis, hired cars, some pick-ups and microcars - and all fuel types count. Second, the reporting basis is documented: figures come from national trade organizations, OICA members or correspondents, national statistics offices or ministries of transport, and where both exist, registration data is preferred over sales data because public authorities publish it.

Datadory delivers this as a production feed: the full 27-year panel of country-by-vehicle-type unit counts, shaped so a single query returns any country's annual output history next to its peers'.

One row per country per period, five columns, nothing nested:

year_period       : 2024
country_region    : China
passenger_cars    : 24380000
commercial_veh    : 3950000
total_vehicles    : 28330000

Illustrative shape - request a sample for verified current-period values. The structure holds everywhere in the archive: a period, a geography, the two vehicle-type splits and their total. Region aggregates such as Africa, Europe and World Total use identical columns, so stacking years into one panel takes no reshaping.

Five fields per row, applied identically to country tables and region aggregates. Definitions follow OICA's own documentation, so the passenger-car column already excludes two-wheelers and anything seating ten-plus without extra filtering logic.

Geography - worldwide by producing country, plus OICA's own regional aggregates (Africa, Europe, World Total among them). One honest caveat travels with the series: the published tables state that data are not available for some European countries, so treat absent rows as missing reports, not zero output.

Temporal - 1999 through 2026. Annual tables form the spine; quarterly releases such as 2026 Q1 give intra-year reads on the same structure, useful when a full calendar year has not closed.

Granularity - country-by-country unit counts per period, split into exactly two vehicle classes plus a total. Roughly 40 countries and regions per annual table across about 27 years of releases. There is no model-level or plant-level depth here - this is the national-output layer, and it pairs naturally with sales-side datasets when you need demand to compare against supply.

API, files, or your warehouse. Daily, weekly, or hourly.

• Market-entry and capacity studies - 27 years of country-level output shows where assembly capacity actually consolidated, which beats consultant-deck assertions when a plant-location case goes to a board.
• Supplier footprint planning - parts and equipment firms locate against production volumes, not registration volumes; the passenger-car versus commercial-vehicle split tells an axle supplier and a cabin-air-filter supplier different stories.
• Economic and trade analysis - national output series feed industrial-production work, import/export context and GDP-nowcast models, with the World Total row serving as a ready denominator.
• Competitive benchmarking - tracking several countries' quarterly releases side by side flags shifts in relative production position within weeks of publication instead of at year-end.

Corporate strategy and market-intelligence teams at OEMs and tier suppliers get the long-run comparative baseline their scenario models assume but rarely have. Economists and policy analysts get a consistently-defined national output series whose reporting basis is documented rather than reverse-engineered. Consultants and market researchers get citable country rankings back to 1999 for automotive industry reports. Data engineers get a five-column flat table that loads without transformation and stacks cleanly across periods.

Three caveats worth having upfront. First, the country coverage moves: the tables carry a standing notice that data are not available for some European countries, so a given year may be short rows for reasons of national reporting rather than production reality. Second, granularity stops at country level - there are no manufacturer, model or plant breakdowns here; if you need those, pair this series with a sales-side dataset rather than expecting it from this one. Third, quarterly releases have appeared in some years and not others, so intra-year analysis should confirm which quarters exist for the period in question before building on them.