World Bank Indicators — GDP, Inflation, Population & 16k More

Country × indicator × year data straight from the World Bank Indicators API v2: GDP, GDP per capita, inflation, population, unemployment, trade, FDI, debt, interest rates, internet users, CO2, poverty… for any of 217 economies and 78 aggregates (World, EU, OECD, income groups, regions). Built for analysts, researchers and AI agents that need clean macro data with units, latest-value tables, YoY change and wide comparison tables. No API key, no proxy, no browser.

Use cases

• Country due diligence — pull GDP, growth, inflation, debt, reserves and FX rate for a market before entering it (indicatorPreset: "finance").
• Market sizing — population, urbanisation, internet users, GDP per capita PPP for a region (region: "ECS", incomeLevel: "UMC").
• Country ranking tables — one row per country with latest value and YoY change (mode: "latest", countries: ["all"]).
• Trade & export research — exports/imports % GDP, merchandise trade, tariffs, FDI for a set of partners as one wide table (indicatorPreset: "trade", wide: true).
• ESG / climate screening — CO2 per capita, renewables share, PM2.5, forest area (indicatorPreset: "energyClimate").
• Monthly macro monitoring — CPI, industrial production, exchange rates from the Global Economic Monitor (source: 15, frequency: "M").
• Indicator discovery for agents — search the 1,500 WDI (or 29,500 total) indicators by text and topic, read definitions (mode: "indicators", mode: "indicatorMeta").

Input

Field	Type	Default	Allowed values / notes
mode	string	data	data time series · latest one row per country×indicator · indicators catalogue search · countries country metadata · sources database list · indicatorMeta definitions
countries	string[]	— (required for data/latest unless a filter is set)	ISO2 (KZ), ISO3 (KAZ), English names (Kazakhstan), all (every economy; aggregates only with includeAggregates: true), aggregate codes (WLD, EUU, OED, HIC, ECS…). Case, accents, spaces and punctuation do not matter. An unknown value → warning + SUMMARY.errors with a "did you mean" (not a dataset row, not billed); the other countries are still read. No known value at all → the run stops with that message
indicators	string[]	macro set	Any World Bank indicator codes. Empty = indicatorPreset
indicatorPreset	string	custom	macro, trade, demographics, finance, digital, energyClimate, social, business, custom (see Reference)
years	integer	5	Most recent N periods (API mrv), 1–60. Not used by mode: latest
startYear / endYear	integer	—	Explicit range (API date=start:end), overrides years/mrnev. With mode: latest: the latest value inside the range
mrnev	integer	—	N most recent non-empty values per country (skips missing years), overrides years. Not used by mode: latest (it always takes latest + previous non-empty value)
gapfill	boolean	false	With years only: API fills missing years with last value
frequency	string	Y	Y, Q, M — sub-annual only for sources 15 (GEM, monthly) and 20/22/23 (quarterly debt)
source	integer	2	World Bank database id (2 = WDI). List with mode: "sources"
includeNulls	boolean	false	Keep rows with value: null
includeChange	boolean	false	Add yoyChangePct, previousYear, previousValue per row (vs previous non-null period)
wide	boolean	false	One row per country × period, indicator codes as columns (saved after the last indicator is read)
fields	string[]	— (all columns)	Keep only these columns, in this order; id is always kept. Columns per mode: see Output. Case, spaces, _ and - do not matter; a name that is not a column of the mode is skipped and named in the status and SUMMARY.warnings; if none is a column, the run stops before any request with the list of valid columns
maxItems	integer	1000	Cap on dataset rows (1–50,000); pagination is automatic and stops once the cap is reached
query	string	—	mode: indicators: text to match in code, name, definition
topic	string	—	mode: indicators: topic id 1–21 (see Reference)
searchAllSources	boolean	false	mode: indicators: search all ~29,500 indicators instead of only source
region	string	—	EAS, ECS, LCN, MEA, NAC, SAS, SSF, NA (aggregates) — selects countries in data/latest modes when countries is empty or all; filters mode: countries
incomeLevel	string	—	HIC, UMC, LMC, LIC
lendingType	string	—	IBD, IDX, IDB, LNX
includeAggregates	boolean	false	Include the aggregates (World, EU, regions, income and lending groups — region "Aggregates" in the World Bank country metadata) in countries: ["all"], mode: countries and region/income selections. Aggregate codes you list explicitly (WLD, EUU…) are always returned. Every data row has isAggregate.

Backward compatible: { "countries": [...], "indicators": [...], "years": N } works exactly as before.

mode, indicatorPreset, frequency, topic, region, incomeLevel and lendingType are fixed lists (values above and in Reference): Apify refuses any other value before the run starts, with the list of allowed values.

Reference

Indicator presets (all codes verified against the API in September 2026)

Preset	Codes
macro (default)	NY.GDP.MKTP.CD GDP US$ · NY.GDP.PCAP.CD GDP per capita · FP.CPI.TOTL.ZG CPI inflation % · SP.POP.TOTL population · SL.UEM.TOTL.ZS unemployment %
trade	NE.EXP.GNFS.ZS NE.IMP.GNFS.ZS NE.TRD.GNFS.ZS (% GDP) · TX.VAL.MRCH.CD.WT TM.VAL.MRCH.CD.WT merchandise US$ · BX.KLT.DINV.CD.WD FDI inflows · BN.CAB.XOKA.GD.ZS current account % GDP · TM.TAX.MRCH.WM.AR.ZS tariff %
demographics	SP.POP.TOTL SP.POP.GROW SP.URB.TOTL.IN.ZS SP.POP.65UP.TO.ZS SP.DYN.LE00.IN SP.DYN.TFRT.IN SM.POP.NETM SL.TLF.TOTL.IN
finance	FR.INR.LEND FR.INR.RINR FM.LBL.BMNY.GD.ZS FS.AST.PRVT.GD.ZS FB.AST.NPER.ZS PA.NUS.FCRF FI.RES.TOTL.CD GC.DOD.TOTL.GD.ZS GC.TAX.TOTL.GD.ZS DT.DOD.DECT.CD
digital	IT.NET.USER.ZS IT.CEL.SETS.P2 IT.NET.BBND.P2 IT.NET.SECR.P6 TX.VAL.TECH.CD GB.XPD.RSDV.GD.ZS
energyClimate	EG.USE.PCAP.KG.OE EG.ELC.ACCS.ZS EG.FEC.RNEW.ZS EG.ELC.RNEW.ZS EN.GHG.CO2.PC.CE.AR5 EN.GHG.CO2.MT.CE.AR5 EN.ATM.PM25.MC.M3 AG.LND.FRST.ZS
social	SE.PRM.ENRR SE.TER.ENRR SE.ADT.LITR.ZS SH.XPD.CHEX.GD.ZS SH.MED.PHYS.ZS SI.POV.DDAY SI.POV.GINI SI.DST.10TH.10
business	IC.BUS.NREG IC.BUS.NDNS.ZS LP.LPI.OVRL.XQ GC.TAX.TOTL.GD.ZS BX.KLT.DINV.WD.GD.ZS FS.AST.PRVT.GD.ZS IP.PAT.RESD VC.IHR.PSRC.P5

The built-in dictionary (src/lib.ts, ~110 codes) also carries a unit for each of these plus GDP PPP, GNI per capita, GDP deflator, sector value added, remittances, external debt ratios, military spending, reserves in months of imports, market cap, youth unemployment, labour participation, infant mortality, tourism, air passengers, container traffic, cereal production/yield, agricultural land, health spending per capita, hospital beds, national poverty line, women in parliament, R&D researchers, patents, journal articles, CPIA ratings. For any other code the unit is parsed from the indicator name ("… (% of GDP)" → % of GDP).

Archived codes that return no data (the actor warns): EN.ATM.CO2E.PC / EN.ATM.CO2E.KT (use EN.GHG.CO2.PC.CE.AR5 / EN.GHG.CO2.MT.CE.AR5), Doing Business IC.BUS.EASE.XQ, IC.REG.DURS, IC.REG.COST.PC.ZS, IC.TAX.TOTL.CP.ZS (discontinued 2021). Sparse series: FR.INR.LEND / FR.INR.RINR / FR.INR.DPST (IMF coverage, no KZ; DPST no US), SI.POV.*, SE.ADT.LITR.ZS, LP.LPI.OVRL.XQ (survey years only) — use mrnev for these.

Regions, income levels, lending types

Region	Income level	Lending type
EAS East Asia & Pacific	HIC High income	IBD IBRD
ECS Europe & Central Asia	UMC Upper middle income	IDX IDA
LCN Latin America & Caribbean	LMC Lower middle income	IDB Blend
MEA Middle East, North Africa, Afghanistan & Pakistan	LIC Low income	LNX Not classified
NAC North America
SAS South Asia
SSF Sub-Saharan Africa

Common aggregate codes (usable in countries)

WLD World · HIC UMC LMC LIC MIC LMY income groups · OED OECD · EUU European Union · EMU Euro area · ARB Arab World · EAS ECS LCN MEA NAC SAS SSF regions · ECA EAP LAC MNA SSA regions excl. high income · CEB Central Europe & Baltics · LDC least developed · HPC HIPC · SST small states · IDA IBD lending groups. All 78 aggregates: {"mode":"countries","region":"NA"}.

Topics (topic for indicator search)

1 Agriculture & Rural Development · 2 Aid Effectiveness · 3 Economy & Growth · 4 Education · 5 Energy & Mining · 6 Environment · 7 Financial Sector · 8 Health · 9 Infrastructure · 10 Social Protection & Labor · 11 Poverty · 12 Private Sector · 13 Public Sector · 14 Science & Technology · 15 Social Development · 16 Urban Development · 17 Gender · 18 MDGs · 19 Climate Change · 20 External Debt · 21 Trade

Most useful sources (source)

id	Database	Frequency
2	World Development Indicators (default)	annual
3	Worldwide Governance Indicators	annual
6	International Debt Statistics	annual
14	Gender Statistics	annual
15	Global Economic Monitor (CPI, IP, FX, stock indices…)	monthly / quarterly
16	Health Nutrition and Population Statistics	annual
20 / 22 / 23	Quarterly Public Sector Debt / External Debt (SDDS, GDDS)	quarterly
27	Global Economic Prospects (forecasts)	annual
28	Global Findex	survey years
31	CPIA	annual
46	Sustainable Development Goals	annual
66	Logistics Performance Index	survey years
75	ESG Data	annual

Full list of 71 databases: {"mode":"sources"}. Indicator codes differ per source — find them with {"mode":"indicators","source":15,"query":"cpi"}.

Examples

Default — macro snapshot of three countries, last 5 years (74 rows)

{ "countries": ["KZ", "US", "CN"] }

Central Asia trade table 2015–2025, one row per country × year (55 rows, 8 indicator columns)

{ "countries": ["KZ", "UZ", "KG", "TJ", "TM"], "indicatorPreset": "trade", "startYear": 2015, "endYear": 2025, "wide": true }

World ranking of GDP per capita with YoY change (~214 economies with data; add "includeAggregates": true for ~47 more rows of World/regional/income-group aggregates, flagged isAggregate: true)

{ "mode": "latest", "countries": ["all"], "indicators": ["NY.GDP.PCAP.CD"], "maxItems": 300 }

Upper-middle-income Europe & Central Asia — digital market sizing, latest non-empty values

{ "mode": "latest", "region": "ECS", "incomeLevel": "UMC", "indicatorPreset": "digital" }

Sparse series done right — poverty & Gini, 3 most recent available observations

{ "countries": ["KZ", "UZ", "IN", "BR"], "indicators": ["SI.POV.GINI", "SI.POV.DDAY"], "mrnev": 3 }

Monthly CPI from the Global Economic Monitor with month-over-month change

{ "countries": ["US", "KZ", "TR"], "indicators": ["CPTOTNSXN"], "source": 15, "frequency": "M", "years": 24, "includeChange": true }

Find indicators about inflation, read one definition

{ "mode": "indicators", "query": "inflation", "maxItems": 20 }
{ "mode": "indicatorMeta", "indicators": ["FP.CPI.TOTL.ZG", "NY.GDP.DEFL.KD.ZG"] }

Output

Data row (mode: data):

{
  "id": "KAZ:NY.GDP.MKTP.CD:2025",
  "country": "Kazakhstan",
  "countryCode": "KAZ",
  "countryIso2": "KZ",
  "indicator": "NY.GDP.MKTP.CD",
  "indicatorName": "GDP (current US$)",
  "year": 2025,
  "period": "2025",
  "value": 306239209650.024,
  "unit": "current US$",
  "obsStatus": null,
  "decimal": 0,
  "source": "World Development Indicators",
  "sourceId": "2",
  "lastUpdated": "2026-07-13",
  "fetchedAt": "2026-09-12T23:40:59.004Z",
  "sourceUrl": "https://api.worldbank.org/v2/country/KAZ;USA;CHN/indicator/NY.GDP.MKTP.CD?format=json&mrv=5&source=2&per_page=1000&page=1"
}

Field	Description
id	ISO3:indicator:period (stable)
country, countryCode, countryIso2	Name, ISO3, ISO2
isAggregate	true for World/regional/income-group/lending-group aggregates (region "Aggregates" in the World Bank country metadata) — filter on it instead of on names
indicator, indicatorName	Code and official name
year, period	Year (number) and raw period (2025, 2026M07, 2024Q3)
value	Number or null (only with includeNulls)
unit	From the built-in dictionary, else parsed from the name, else null
obsStatus, decimal	API observation status (e.g. E estimate) and suggested decimals
source, sourceId, lastUpdated	Database name/id and its last update date
yoyChangePct, previousYear, previousValue	With includeChange — % change vs previous non-null period of the same series
fetchedAt, sourceUrl	Fetch time and the exact API URL

mode: latest rows: country, countryCode, countryIso2, indicator, indicatorName, unit, latestYear, latestPeriod, latestValue, previousYear, previousValue, yoyChangePct, source, isAggregate, fetchedAt, sourceUrl. wide: true rows: id (ISO3:period), country, countryCode, countryIso2, year, period, isAggregate, <indicator code>: value…, fetchedAt, sourceUrl. mode: indicators / indicatorMeta: indicator, indicatorName, unit, sourceId, source, topics, topicIds, sourceNote, sourceOrganization, inDictionary (+ dictionaryNote). mode: countries: countryCode, countryIso2, country, isAggregate, region, regionName, adminRegion, incomeLevel, incomeLevelName, lendingType, lendingTypeName, capitalCity, latitude, longitude. mode: sources: sourceId, code, name, lastUpdated, dataAvailability, metadataAvailability. Every row also has id, fetchedAt and sourceUrl. With fields, a row keeps id plus the listed columns in your order, e.g. "fields": ["country", "year", "value"] → { "id": …, "country": …, "year": …, "value": … }.

Problems are not dataset rows (every row is charged): the SUMMARY record in the key-value store lists errors ([{ "input": "FOO.BAR", "error": "World Bank API error 120 Invalid value: …" }], { "input": "Kazakstan", "error": "Unknown country \"Kazakstan\" (did you mean KAZ Kazakhstan?) …" }), errorCount, notFound (indicators with no data for the selection, unknown indicator codes in indicatorMeta), notFoundCount, indicatorsNotRead / indicatorsReadInPart (a stop at maxItems, at your spending limit or before the run timeout), warnings and the run counts (delivered, stoppedBy). The status message sums them up and each is logged as a warning. The run does not fail because of one bad code — only when nothing could be fetched at all.

Use it from code / agents

curl -X POST "https://api.apify.com/v2/acts/yadroo~world-bank-indicators/run-sync-get-dataset-items?token=$APIFY_TOKEN" \
  -H 'Content-Type: application/json' \
  -d '{"mode":"latest","countries":["KZ","UZ","KG"],"indicatorPreset":"macro"}'

import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('yadroo/world-bank-indicators').call({
    countries: ['KZ', 'US', 'CN'], indicators: ['NY.GDP.MKTP.KD.ZG'], startYear: 2010, endYear: 2025, includeChange: true,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

from apify_client import ApifyClient
client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("yadroo/world-bank-indicators").call(run_input={
    "mode": "latest", "countries": ["all"], "indicators": ["SP.POP.TOTL"], "maxItems": 300})
items = client.dataset(run["defaultDatasetId"]).list_items().items

MCP: add https://mcp.apify.com to your agent (Claude, Cursor, custom) and call the yadroo/world-bank-indicators tool with the same JSON input — the input schema descriptions are written for agents.

Pricing

Pay-per-event: $0.001 per run start + $0.0005 per dataset item.

Run	Items	Cost
Default (3 countries × 5 indicators × 5 years)	74	≈ $0.038
Trade table, 5 countries × 11 years, wide	55	≈ $0.029
World ranking, 1 indicator, latest	~214	≈ $0.108
All countries × 5 indicators × 5 years	~6,600	≈ $3.30

Store discounts: Bronze −10 %, Silver −20 %, Gold and above −30 % on the row price; the start event is the same on every plan; platform usage is included. Unknown codes and indicators without data are never rows, so they cost nothing. Use wide/latest to get more information per row.

Your Maximum cost per run is respected: the run stops at it and ends with "Stopped at your spending limit: N rows delivered" (the indicators left unread are named in SUMMARY).

Limits & FAQ

• Rate limits / politeness — one request per indicator (per 60-country batch, per 1,000-row page) with a 250 ms pause, up to 3 attempts with backoff on 429/5xx/timeouts (Retry-After honoured). The World Bank API is free and public; a typical run is 1–15 requests.
• Freshness — WDI is refreshed a few times a year (the lastUpdated field tells you when); GEM (source 15) monthly. Data is not real time — for FX use our fx-rates actor, for yields us-treasury-yields.
• Empty results — nulls are dropped by default, so a country with no data simply produces no rows; an indicator with no rows at all is listed in SUMMARY.notFound (free, not a dataset row). Use mrnev for sparse series, includeNulls: true to see the gaps.
• Invalid codes — invalid indicator → the API error is logged and listed in SUMMARY.errors; unknown country → validated locally against the live country list (the API would silently drop it) and reported the same way with a "did you mean" (never swapped for another country). Neither costs anything.
• years vs startYear vs mrnev — the API accepts only one; the actor picks range > mrnev > years and warns about the ignored ones. mode: latest takes the latest non-empty value of each series (with startYear/endYear: the latest inside that range) and warns when years or mrnev are set.
• maxItems — counts final dataset rows. Rows are saved as they arrive (per API page in mode: data, per indicator in mode: latest or with includeChange), and reading stops once maxItems rows are saved — nothing is fetched ahead; the indicators left unread are named in the status and SUMMARY.indicatorsNotRead. wide tables need every indicator, so they are saved after the last one.
• Run timeout — a run near its timeout stops starting new requests, saves what it has read and ends SUCCEEDED with "Stopped before the run timeout: N rows saved". A typical run takes seconds; all × 10 indicators × 60 years is ~170 requests (a few minutes).
• Memory — 256 MB is enough; all × many indicators × 60 years can be tens of thousands of rows — raise maxItems and expect a longer run.
• Licence — World Bank Open Data, CC BY 4.0; cite "World Bank, World Development Indicators".
• Roadmap — per-capita/ratio calculators across two indicators, CAGR over the range, CSV-friendly column naming for wide.

---

Made by Yadroo. Pairs with fx-rates, us-treasury-yields, sec-company-financials, sec-edgar-filings, public-holidays, openalex-works and wikipedia-search.