🌐 IMF Indicator Catalog
April 25, 2026 · View on GitHub
Canonical catalog of IMF indicators relevant to European Parliament intelligence, organised by policy domain. Complements
analysis/worldbank/indicator-catalog.md; together they cover the full economic/social/environmental surface required by policy article types.
📅 Last Updated: 2026-04-25 | 🏷️ Classification: Public | 🌀 Wave: 4
1. Database Inventory (policy-relevant subset)
| Database | Name | Frequency | Forecasts | Typical EP use |
|---|---|---|---|---|
WEO | World Economic Outlook | Annual (+Q in some series) | ✅ +5y | ECON/BUDG/AFET macro/fiscal backdrop |
FM | Fiscal Monitor | Annual | ✅ +5y | BUDG debt/deficit, primary/structural balance |
GFSR | Global Financial Stability Report | Semi-annual (Apr + Oct) | — | ECON financial-stability commentary |
EREO | Regional Economic Outlook — Europe | Semi-annual (May + Nov) | ✅ +3y | ECON/AFET Europe-specific narrative |
IFS | International Financial Statistics | Monthly / Quarterly | — | ECON policy-rate tracking, quarterly growth context |
CPI | Consumer Price Index | Monthly | — | Breaking news inflation context |
BOP_AGG | Balance of Payments — Aggregates | Quarterly | — | INTA trade balance, FDI flow monitoring |
BOP | Balance of Payments — detailed components | Quarterly | — | INTA component drill-down |
DOT | Direction of Trade Statistics | Monthly / Quarterly | — | INTA bilateral trade flows EU ↔ RoW |
ER | Exchange Rates (REER / NEER) | Monthly | — | ECON FX-competitiveness framing |
PCPS | Primary Commodity Price System | Monthly | — | ITRE energy, AGRI food price context |
CDIS | Coordinated Direct Investment Survey | Annual | — | INTA long-run bilateral FDI stock |
CPIS | Coordinated Portfolio Investment Survey | Semi-annual | — | ECON cross-border portfolio holdings |
IRFCL | International Reserves & FX Liquidity | Monthly | — | ECON monetary-policy & reserve commentary |
SRF | Standardized Report Forms (monetary) | Monthly | — | ECON central-bank balance sheet |
GFS | Government Finance Statistics | Annual (+Q subset) | — | BUDG revenue structure, ECON fiscal stance |
FSI | Financial Soundness Indicators | Quarterly | — | ECON banking-sector stability |
HSG | Historical Spending Dataset | Episodic | — | Long-run fiscal context (historical-baseline) |
HPDD | Historical Public Debt Database | Episodic | — | Long-run debt-to-GDP comparisons |
HISTINF | Historical Inflation Dataset | Episodic | — | Long-run inflation context |
SDG | Sustainable Development Goals (IMF) | Annual | — | ENVI / DEVE SDG monitoring |
The full upstream inventory includes ~155 databases — see
database-directory.md for the complete
editorial-relevance map. Use imf-search-databases to discover any
database not pre-listed.
2. Policy Domain → IMF Indicator Mapping
2.1 Macroeconomic / ECON
| Indicator key | Database | SDMX code | Freq | Forecast | Display label |
|---|---|---|---|---|---|
gdp | WEO | NGDPD | A | ✅ | GDP, current USD |
gdpGrowth | WEO | NGDP_RPCH | A | ✅ | Real GDP growth |
gdpPerCapita | WEO | NGDPDPC | A | ✅ | GDP per capita, current USD |
inflation | WEO | PCPIPCH | A | ✅ | CPI inflation |
unemployment | WEO | LUR | A | ✅ | Unemployment rate |
currentAccount | WEO | BCA_NGDPD | A | ✅ | Current account, % of GDP |
policyRate | IFS | FPOLM_PA | M | — | Monetary policy rate |
realEffectiveExchangeRate | ER | EREER_IX | M | — | Real effective exchange rate |
2.2 Fiscal / BUDG
| Indicator key | Database | SDMX code | Freq | Forecast | Display label |
|---|---|---|---|---|---|
govDebt | FM | GGXWDG_NGDP | A | ✅ | General government gross debt, % of GDP |
primaryBalance | FM | GGXONLB_NGDP | A | ✅ | Primary balance, % of GDP |
structuralBalance | FM | GGSB_NPGDP | A | ✅ | Structural balance, % of potential GDP |
2.3 Trade / INTA
| Indicator key | Database | SDMX code | Freq | Forecast | Display label |
|---|---|---|---|---|---|
exportsGdp | WEO | TX_RPCH | A | ✅ | Export volume growth |
fdiInflow | BOP_AGG | BFD_BP6_USD | Q | — | FDI inflow, USD |
2.4 Defence / AFET / SEDE
The WEO does not publish MS.MIL.XPND.GD.ZS directly; use World Bank for
military-expenditure-to-GDP and cross-reference with IMF general-government
expenditure aggregates where needed.
2.5 Not covered by IMF — remain on World Bank
- Social (population, life expectancy, internet users) — WB
SP.POP.*,SP.DYN.LE00.IN,IT.NET.USER.ZS - Health (physicians, hospital beds, immunisation) — WB
SH.MED.*,SH.IMM.* - Education (literacy, enrolment, teachers) — WB
SE.* - Environment (CO₂, renewable energy) — WB
EN.ATM.CO2E.PC,EG.FEC.RNEW.ZS - Innovation (R&D, high-tech exports) — WB
GB.XPD.RSDV.GD.ZS,TX.VAL.TECH.MF.ZS
3. Frequency & Release Cadence
| Dataset | Release cadence | Typical lag (actuals) | Forecast horizon |
|---|---|---|---|
| WEO | April + October | 3 months | Current + 5 years |
| FM (Fiscal Monitor) | April + October | 3 months | Current + 5 years |
| IFS | Monthly auto-publication (SDMX 3.0) | 4–8 weeks (Q) | — |
| CPI | Monthly | 4–6 weeks | — |
| BOP_AGG | Quarterly | ~10–13 weeks | — |
| ER | Monthly | ~2 weeks | — |
| PCPS | Monthly | ~2 weeks | — |
4. Observation Attributes
IMF SDMX-JSON responses may carry OBS_STATUS with the following codes
relevant to EU Parliament Monitor:
| Code | Meaning | EP handling |
|---|---|---|
A | Actual | Default; no marker required |
F | Forecast / Projection | MUST be labelled "forecast" or "projection" in prose, with vintage |
E | Estimate | Labelled "IMF estimate" with vintage |
B | Break in series | Cite methodological footnote |
P | Provisional | Treated like A with "provisional" prose label |
The native TypeScript client in src/mcp/imf-mcp-client.ts
returns the raw SDMX-JSON payload inside the MCPToolResult envelope
(response.content[0].text). Agents parse it with any standard
SDMX-JSON reader. This catalog does not imply that the aggregator or
client automatically propagates OBS_STATUS into rendered HTML. If a row
represents an F (forecast) observation, the authoring/template layer
MUST add data-forecast="true" explicitly to that table row after
inspecting the source data. That explicit markup enables the Chart.js and
CSS hooks described in
chart-integration-guide.md §7.
5. Confidence & Methodological Caveats
- IMF optimism bias: WEO forecasts historically overstate medium-term growth (see IMF's own 2015 Forecast Accuracy Review). Articles citing forecasts SHOULD include one sentence acknowledging the methodological caveat, especially for 3+ year horizons.
- Vintage drift: Each WEO vintage revises prior forecasts. Record
vintage in the article HTML (
data-vintage="WEO-April-2026") so reprocessing surfaces the original data-as-of time. - Membership aggregates:
EA(Euro Area) membership historically grew (Croatia joined in 2023); timeseries use the current membership, not the contemporaneous one. Cite "EA current membership" in charts. - Country code drift: Kosovo is
UVKat IMF on some legacy datasets vs.XKXat WB. Seeeu-country-mapping.md §4.
6. Fingerprint Convention
Stage-C editorial review uses the following SDMX-code and product-name strings as the authoritative fingerprints for an "IMF citation" in any article:
- Product names:
IMF,International Monetary Fund,WEO,World Economic Outlook,Fiscal Monitor,data.imf.org,dataservices.imf.org. - Virtual tool names (the five exposed by
src/mcp/imf-mcp-client.ts):imf-list-databases,imf-search-databases,imf-get-parameter-defs,imf-get-parameter-codes,imf-fetch-data. - SDMX indicator codes: every code in Section 2 above
(
NGDPD,NGDP_RPCH,PCPIPCH,LUR,GGXWDG_NGDP,GGXONLB_NGDP,FPOLM_PA,BFD_BP6_USD,EREER_IX, …).
Drift guard: test/integration/mcp/imf-mcp.test.js
fails if the five tool names in IMF_MCP_TOOLS (exported from
src/mcp/imf-mcp-client.ts) diverge
from the list above.
The earlier runtime fingerprint tables (
IMF_STRONG_FINGERPRINTS,IMF_INDICATOR_CODES,IMF_INDICATOR_SDMX_CODES,IMF_POLICY_INDICATORS) lived insrc/utils/content-validator.tsandsrc/utils/imf-data.tsand were purged in the April-2026 aggregator-pipeline migration. Fingerprinting is now an editorial responsibility enforced at Stage C per.github/prompts/04-article-generation.md §5.