Every year someone asks where to actually get free economic data, and every year the honest answer is "it depends what you need," not a single winner. Here's the real list, including where each option's limits are.

## FRED — best for U.S. macro series

FRED (Federal Reserve Economic Data), maintained by the Federal Reserve Bank of St. Louis, is free and has been a default reference for U.S. macro data for years. Its historical depth on flagship U.S. series is excellent, and there's no paid tier to work around. The limit is scope: international coverage is comparatively thin, and there's no bilateral trade data by product and partner.

## World Bank WDI — the broadest country coverage

The World Bank's World Development Indicators is free, open, and covers close to every World Bank member economy — the broadest country list of any source here. It's authoritative well beyond core macro data, into poverty, health, and education indicators. The tradeoff is update cadence: many WDI indicators refresh annually, which is too slow for anyone tracking a monthly or quarterly move.

## IMF — the backbone for GDP, prices, and government finance

The IMF publishes GDP growth, inflation, unemployment, and government finance data through its SDMX data services, and it's the source behind a large share of the indicators on our own [country pages](/germany) and [indicator hubs](/indicators/gdp). It's free to query directly, though the raw SDMX interface is more work to integrate against than a REST API built specifically for application use.

## BIS — for central bank policy rates

The Bank for International Settlements aggregates central bank policy rates and official exchange rates across a wide set of countries — a narrower but useful free source if that's specifically what you need, rather than a full macro dataset.

## Where EconomicsAPI fits

We're not a fifth primary source competing with the above — we standardize the same class of official data (IMF, BIS, World Bank WDI, and select FRED series for U.S. monthly data) into one consistent, [daily-checked](/blog/how-we-source-and-verify-economic-data) API and MCP server, and add bilateral trade data by product and partner that none of the above provide. The free tier covers the [full 50-country dataset](/economy) and every endpoint, no credit card required. If your need is a single well-known U.S. series, FRED alone may be all you need. If you need multiple countries, [trade data](/germany/exports), and structured [agent access](/ai) in one place, that's the specific gap we built for — see the [methodology](/methodology) for exactly how we source and verify what we publish.

## The practical recommendation

Start with FRED for U.S.-only questions, the World Bank for the broadest country reach on slower-moving indicators, and the IMF directly if you're comfortable with SDMX and need government finance or external-sector data. Reach for a standardized API — ours or another — once you need multiple countries, faster update checking, or trade data in the same schema as your macro indicators. All of these are free; the decision is about fit, not cost.

A practical, honest guide to FRED, the World Bank, the IMF, and where EconomicsAPI fits alongside them — not a sales pitch dressed as a list.

Where to Get Free Economic Data in 2026

Ten years of a single country's trade data tells a more concrete story than most macro headlines manage. Here's what a decade of Germany's bilateral trade flows actually shows, pulled directly from the [Germany trade data](/germany/exports) we publish — no modeling, no forecast, just what got recorded.

## The headline: exports up roughly 35% over the decade

German exports grew from $1.29 trillion in 2015 to $1.75 trillion in 2025 — a cumulative increase of roughly 35%, though not a straight line. Imports grew faster in relative terms over the same period, from $0.99 trillion to $1.52 trillion, meaning Germany's trade surplus narrowed in relative terms even as both sides of the ledger grew in absolute dollars. The 2025 surplus sits at roughly $226 billion. You can see the full year-by-year series, exports against imports, on the [Germany trade hub](/germany).

## The one clear break: 2019 to 2021

The decade isn't a smooth climb. Exports peaked at $1.53 trillion in 2018, slipped to $1.46 trillion in 2019, then dropped further to $1.36 trillion in 2020 as pandemic-era disruption hit global trade broadly. The recovery was sharp: $1.62 trillion in 2021, back above the pre-dip level and continuing to $1.67 trillion by 2022. It's the only multi-year contraction in the ten-year window — everything else is year-over-year growth, some years faster than others.

## What Germany actually exports

By value, passenger vehicles remain Germany's largest single export line — just over 10% of total export value in the latest full year, ahead of pharmaceuticals (medicaments for retail sale, at roughly 4.2%) and vehicle parts (roughly 3.6%). That's consistent with Germany's long-standing reputation as an automotive exporter, but the [full product breakdown](/germany/exports) shows the tail is long: no single product category outside vehicles clears 5% of total export value, which is itself informative about how diversified the remaining export base is.

The fastest-growing export lines tell a different story than the largest ones — passenger and cargo vessels grew roughly 75% year-over-year in the latest reading, and precious metal scrap and silver both grew more than 50%, though all three are small relative to the vehicle trade in absolute dollar terms.

## Who Germany trades with

The Netherlands is Germany's largest trade partner by value, at just over 10% of total trade flow, followed by the United States (roughly 8.2%), Poland (roughly 6.7%), France (roughly 6.7%), and China (roughly 6.1%). Trade with Poland grew faster year-over-year than trade with any of the other top five partners in the latest reading — a data point worth watching if it continues, though one year isn't a trend. The [Germany trade-partners page](/germany/trade-partners) breaks down the full ranked list.

## Why this level of detail matters

A single "Germany's exports grew 35% over ten years" headline hides all of this — the pandemic-era dip and recovery, which products actually drive the number, which partners are gaining share. That's the case for [bilateral trade data by product and partner](/methodology) rather than a single aggregate figure: the aggregate is directionally right but the detail is where the actual signal lives. The same shape of data — [product and partner breakdowns](/glossary/hs-code), a running [trade balance](/glossary/trade-balance) — is available for every one of the 41 countries we cover, not just Germany, through the same [API](/blog/get-gdp-for-any-country-in-one-api-request) and MCP access described elsewhere on this blog.

German exports grew from $1.29 trillion in 2015 to $1.75 trillion in 2025 — here's what the decade of bilateral trade data actually shows.

Ten Years of German Trade Data: What Changed

An AI agent answering "what's Germany's current inflation rate" has three options: guess from training data that's already stale, scrape a web page and hope the table parses cleanly, or call a tool built for exactly this. MCP — the Model Context Protocol — is what makes the third option as easy as the first two, and it's why we built a hosted MCP server alongside the REST API rather than treating agents as an afterthought.

## What MCP actually gives an agent

MCP standardizes how a model discovers and invokes external tools. Instead of an agent trying to parse the [Germany hub](/germany) page's HTML — fragile, and liable to break the next time we change a layout — it calls a tool that returns the same structured JSON a script would get from `/v1/latest`. Our [AI and agent access policy](/ai) states this directly: citation from rendered pages is welcome, but programmatic access should go through the API or MCP, not scraping.

## One tool per endpoint

Our MCP server mirrors the REST API one-to-one: one MCP tool per [endpoint](/methodology), same parameters, same response shape. An agent that can call `getLatest` over MCP is calling the identical logic behind `GET /v1/latest?country=germany&indicator=gdp` — see [getting GDP for any country in one request](/blog/get-gdp-for-any-country-in-one-api-request) for that call worked through directly against the REST API, which is useful context even if your agent will ultimately call it over MCP instead.

## What's covered

Coverage over MCP is identical to the REST API: [50 countries](/economy) across [20 economic indicators](/indicators) — GDP, inflation, unemployment, policy rates, government finances, and more — plus bilateral trade data by product and partner for 41 of those countries, like [Germany's export mix](/germany/exports). Nothing is held back for one access method over the other.

## A typical agent workflow

A research agent tasked with "compare inflation across the G20 this quarter" can call the MCP equivalent of `/v1/latest?country=g20&indicator=inflation-rate` in one tool call, get back 20 source-attributed values, and reason over structured numbers instead of trying to extract them from prose. That same query, run manually, is the [inflation indicator hub](/indicators/inflation-rate) sorted by country — useful for a person, but the point of MCP is that an agent doesn't have to read it that way.

## Getting connected

MCP access is free, same as the REST API — no separate license, no sales conversation. Read the AI and agent access policy linked above in full for what agents may and may not do with this data — citation, bulk extraction, training use — before you point a crawler or an agent at the site.

How to connect an AI agent to real GDP, inflation, and trade data over MCP instead of scraping pages or hand-rolling API calls.

Giving Your AI Agent Live Economics Data via MCP

The fastest way to understand an API is a request that actually works. Here's GDP for Germany, end to end, against the real production endpoint — no placeholder values.

## The request

```http
GET https://api.economicsapi.com/v1/latest?country=germany&indicator=gdp
Authorization: Bearer sk_live_xxxxxxxxxxxxxxxxxxxx
```

`country` accepts an ISO3 code (`DEU`), the slug (`germany`), or a group token like `g20` or `eu27` — see the [full country list](/economy) for every slug we support. `indicator` accepts the KPI ID or the slug; `gdp` is one of the [20 indicators](/indicators) covered for every country that carries it.

## The response

```json
{
  "data": [
    {
      "series_id": "DEU.NA.GDP.CUR.USD",
      "country_id": "DEU",
      "country": "Germany",
      "kpi_id": "NA.GDP.CUR.USD",
      "indicator": "GDP, current prices, USD",
      "slug": "gdp",
      "group": "gdp",
      "latest_value": 5050922925047.05,
      "latest_value_date": "2025-12-31",
      "previous_value": 4685592577804.69,
      "previous_value_date": "2024-12-31",
      "highest_value": 5050922925047.05,
      "highest_value_date": "2025-12-31",
      "lowest_value": 84619845881.26,
      "lowest_value_date": "1960-12-31",
      "unit": "USD bn",
      "frequency": "annual",
      "source_name": "World Bank",
      "source_url": "https://data.worldbank.org"
    }
  ]
}
```

That's the same figure shown on the [Germany hub](/germany) and the [Germany GDP page](/germany/gdp) — this is a live production endpoint, not a demo dataset with different numbers than the site.

## Multiple countries in one call

Comma-list the `country` parameter to avoid one request per country:

```http
GET https://api.economicsapi.com/v1/latest?country=germany,france,japan&indicator=gdp
Authorization: Bearer sk_live_xxxxxxxxxxxxxxxxxxxx
```

Or use a group token to get an entire cohort — `g20` for the G20, `eu27` for every EU member state — in a single request rather than looping through 20+ separate calls.

## Reading the response fields

`latest_value` and `latest_value_date` are the current reading. `previous_value`, `highest_value`, and `lowest_value` give you trend context — up, down, near a historical extreme — without a second request. `source_name` and `source_url` are the attribution: here, [World Bank](/methodology) World Development Indicators, the same source cited on every GDP-related page on this site. The full field reference, including the other 19 indicators (inflation, unemployment, policy rates, and more), is in the [indicator catalog](/indicators).

## Beyond GDP: history and other indicators

Swap `indicator=gdp` for any other slug — `inflation-rate`, `unemployment-rate`, `government-debt-to-gdp` — against the same `/v1/latest` shape, or point the same country and indicator parameters at `/v1/history` for the full time series instead of just the latest reading. If you're wiring this up for an AI agent rather than a script, the same data is available as an MCP tool — see [giving your agent live economics data](/blog/giving-your-ai-agent-live-economics-data-mcp-guide) for that version of this walkthrough. Every field returned here traces back to a documented source — see [how we source and verify the data](/blog/how-we-source-and-verify-economic-data) for the full picture.

A working example against the real /v1/latest endpoint — one request, any of 50 countries, with source attribution in the response.

Get GDP for Any Country in One API Request

TradingEconomics has been a default reference for economic data and forecasts in financial media for years, and it's earned that reputation — broad coverage, long history, a name analysts recognize. We get asked often enough how EconomicsAPI compares that it's worth writing down plainly, including the parts where the honest answer is "use the other one."

## Where TradingEconomics is genuinely strong

TradingEconomics' scope goes well beyond macro indicators: equities, bonds, commodities, and forecasts sit alongside its economic data, all under one widely cited brand. If your team already builds on that breadth — or needs markets data we simply don't carry — that's a real reason to stay. We don't try to match that scope, and we're not trying to.

## Where the two products diverge

The differences that actually matter for a buyer come down to access, cadence, and how machine-readable the data is meant to be:

**Access.** Full TradingEconomics API access sits behind a sales conversation. Every [endpoint](/methodology) on EconomicsAPI, and the full 50-country dataset behind it, is free with no credit card and no call required.

**Update cadence.** We check IMF and BIS-sourced series daily and World Bank-sourced series weekly against the primary source, and only publish a value when it actually changed — the mechanics are laid out in [how we source and verify data](/blog/how-we-source-and-verify-economic-data). TradingEconomics doesn't publicly document its refresh cadence in the same detail.

**Agent access.** We ship a hosted MCP server — one tool per REST endpoint — so an AI agent can query [GDP](/indicators/gdp), [inflation](/indicators/inflation-rate), or [trade flows](/germany/exports) directly instead of scraping a page. TradingEconomics has no publicly documented MCP server as of September 2026. Our [AI and agent access policy](/ai) covers exactly what's allowed and how agents should connect.

**Scope.** We're economics and bilateral trade only — GDP, prices, labour, money, government, and external-sector indicators for [50 countries](/economy), plus product-and-partner trade data for 41 of them. TradingEconomics covers all of that and markets data besides.

## A worked example

Say you need Germany's current GDP, inflation, and export mix in one place, structured for a downstream tool. On EconomicsAPI that's a `GET` to `/v1/latest?country=germany` for the macro figures, or a visit to the [Germany hub](/germany) for the human-readable version, with sourcing cited inline. TradingEconomics can answer the same question through its own interface, but reaching it programmatically at scale is the piece that requires a sales conversation.

## The honest recommendation

If markets data, forecasts, or TradingEconomics' specific brand recognition in your reporting matters, keep using it — possibly alongside us for the macro and trade slice. If you specifically need free, self-serve access to core economic indicators and bilateral trade data with an MCP server built in from day one, that's the gap we built EconomicsAPI to fill.

An honest comparison of coverage, update cadence, pricing transparency, and agent support — including where TradingEconomics is still the better fit.

TradingEconomics API vs EconomicsAPI, Compared

Ask any team that has tried to build a dashboard on public economic data and you'll hear the same complaint: not that the data is hard to find, but that it's hard to trust. A GDP figure quoted in one place doesn't match another. An indicator hasn't updated in months and nobody can tell if that's because nothing changed or because nobody's watching it. We built EconomicsAPI to remove that ambiguity, and this is the honest version of how.

## The four sources, and why each one

We don't run our own surveys or maintain a proprietary panel. Every number sold as data on this site is standardized from a small set of primary official sources:

- **The International Monetary Fund (IMF)** — the backbone for GDP growth, inflation, unemployment, government finances, and external-sector series across most of the [50 countries](/economy) we cover, published through IMF's SDMX data services.
- **The Bank for International Settlements (BIS)** — the source for central bank policy rates and official exchange rates, aggregated from the world's central banks.
- **The World Bank's World Development Indicators (WDI)** — the source for GDP levels, population, FDI, remittances, and other indicators the IMF doesn't carry for every country.
- **FRED (Federal Reserve Economic Data)**, maintained by the Federal Reserve Bank of St. Louis — used for select U.S. monthly series where it publishes faster than the alternatives.

You can see exactly which source backs a given number on any [indicator page](/indicators) or [country page](/germany) — the attribution is printed next to the value, not buried in a footnote.

## Checked daily, published only when something moves

The IMF and BIS feeds are checked once a day; the World Bank WDI feed, which updates far less frequently at the source, is checked weekly. In both cases the check compares the freshly fetched value against what we already have on file, and a value only moves in our system when the underlying source value has actually changed. That distinction matters more than it sounds: a naive "refresh timestamp" approach makes every series look freshly updated even on days when the source published nothing new, which quietly erodes trust the first time someone notices. This is described in more detail on the [methodology page](/methodology) — and you can see it in practice as the "last updated" date on every country and [indicator](/indicators/gdp) page, which reflects real changes only.

## Coverage: honest about the edges

As of September 2026 we cover 50 countries — the G20 plus every EU-27 state plus the other major economies that also carry [bilateral trade data](/germany/exports) — across 20 indicators spanning GDP, prices, labour markets, money and banking, government finances, and the external sector. That's a deliberately bounded scope. We'd rather ship 50 countries checked daily than 200 countries nobody's verified feels current. If your use case needs a country or indicator we don't yet carry, [get in touch](/contact) — coverage gaps are itemized, not hidden.

## One narrow exception, disclosed

There's exactly one series where we compute rather than pass through: real GDP growth year-over-year, derived from the official constant-price GDP level series, because several sources publish the level but not the growth rate directly. It's the same arithmetic any analyst would do by hand, applied consistently — not a model, not a forecast, and it's the only indicator on the site marked as derived.

## Why this matters if you're evaluating alternatives

If you're comparing providers — including TradingEconomics or going straight to FRED or the World Bank's own APIs — the sourcing question is the one worth asking hardest. Ask any provider, us included, which primary sources back a specific number and how often it's actually checked. If the answer is vague, that's the answer.

We keep every value auditable back to its official source, and the same source-attributed data is available to AI agents over MCP — see our [AI and agent access policy](/ai) for how that works without scraping the site.

Every figure traces to IMF, BIS, World Bank WDI, or FRED — checked daily and published only when a value actually changes.

How We Source and Verify Economic Data

EconomicsAPI

Trade flows and economic data for AI agents

{subject} runs on the same infrastructure whether you're sending your first call or your millionth.

Built for production from day one

No contracts, no monthly minimum, no surprise invoices — start free and scale as you grow.

Pay for what you use, nothing else

Public docs, SDKs, and a hosted MCP server mean your team doesn't wait on us to get started.

Ship your first call in an afternoon

Uptime in the last 12 months, SLO measured.

Free

No credit card required.

Usage-based

From a $50 minimum deposit.

Enterprise

For committed volume.

100 API calls free. Then pay only for what you use.

United States

China

Euro Area

Germany

Japan

United Kingdom

India

France

Russia

Italy

Canada

Brazil

Spain

South Korea

Mexico

Australia

Türkiye

Indonesia

Netherlands

Saudi Arabia

Switzerland

Poland

Belgium

Ireland

Argentina

Sweden

Singapore

Austria

Thailand

Norway

Denmark

Romania

Hong Kong

South Africa

Czechia

Portugal

Finland

Greece

New Zealand

Hungary

Slovakia

Bulgaria

Croatia

Luxembourg

Lithuania

Slovenia

Latvia

Estonia

Cyprus

Malta

The Bloomberg Terminal alternative: no seat, no subscription — just econ and trade data.

A Bloomberg Terminal is reportedly ~$30,000 a seat per year. Ours is usage-based: {free_calls} free calls, then pay per call — no seat, no subscription.

Bloomberg Terminal Alternative: No Seat

Bloomberg Terminal

Public pricing, line by line.

We don't sell seats or annual plans. {free_calls} free calls a month, then pay per call from a ${min_deposit} deposit — that's the whole model.

No seat, no subscription

GDP, inflation, labour, rates, and debt, plus bilateral trade by product and partner — same key, same schema, 2015 to the present.

Econ to trade, any country

REST and a hosted MCP server. No terminal to install, no UI to log into, no provisioning call.

Built for software

Cost

Getting the data

Depth per country

Bloomberg Terminal vs {brand}, line by line.

The Sugra alternative: usage-based, no subscription — and econ-to-trade depth on any country.

Sugra is a broad multi-domain API with volume-based plans. Ours is usage-based: {free_calls} free calls, then pay per call — econ and trade depth on any country.

Sugra Alternative: Usage-Based Econ + Trade

Sugra

{free_calls} free API calls every month, then pay per call from a ${min_deposit} deposit. No subscription and no plan to pick — you pay for the calls you make.

Usage-based, no subscription

Macro indicators and bilateral trade by product and partner, in the same API and MCP server, with one schema across every country we cover.

Econ to trade, one API

50 countries of economics and 41 countries of monthly bilateral trade from 2015 to the present — not a thin slice of a much wider catalog.

Go deep on any country

How you pay

Scope

Access

Where they fit better

Sugra vs {brand}, line by line.

The Trading Economics alternative: usage-based, no subscription — and econ-to-trade depth on any country.

Trading Economics API starts at $149/mo billed yearly for 500 requests. Ours is usage-based: {free_calls} free calls, then pay per call — no subscription.

Trading Economics Alternative: No Subscription

Trading Economics

{free_calls} free API calls every month, then pay per call from a ${min_deposit} deposit. No monthly minimum, no yearly plan. Trading Economics' API starts at $149/month billed yearly for 500 requests.

Go deep on any country we cover: GDP, inflation, labour, rates, and debt, plus bilateral trade by product and partner — same key, same schema, same response shape.

Econ to trade, one place

A hosted MCP server and REST API from one schema, with llms.txt, .md twins, and a machine-readable server card. Use it from Claude, ChatGPT, or Cursor.

Built for agents

Trading Economics vs {brand}, line by line.

Blog