How every module is built

How the FinObservatory crisis layer is built: five primary chronologies, each source's own crisis definitions, dedup rules, cross-source agreement, the two-panel run-up event study (advanced economies and emerging markets), and caveats.

Exact construction of the /crises/r-zone backtest and live reading: variables, cutoffs, panels, chronologies, the unrated rule, and every labeled deviation from Greenwood, Hanson, Shleifer and Sorensen (2022).

The three competing haircut measures, which column is which and how the data proves it, the coverage of the 1823-2024 record by era, and what the haircut data cannot tell you.

How the /housing event study is built: the BIS unit and value_type filter that must be applied to both columns, the annual aggregation rule, the -5/+5 completeness rule that selects the sample, the crisis chronology used, and every limitation.

How the registered-investment-adviser module is built from the SEC Form ADV FOIA archive: why filings are deduplicated to one adviser-year per CRD before any aggregate, how RAUM coverage is handled, and what the employees and clients fields can and cannot say.

How the FinObservatory Composite bank-health score is built: formula, component weights, peer groups, the honest 2023-failures backtest, and the mandatory not-CAMELS caveats.

Deposit-beta construction, fixed joint-distribution bins, the exact 2023Q3 uninsured-funding join, counted coverage, and external validation anchors.

Field-by-field MDRM mapping for the rate-shock fragility map and the CRE concentration screen, the 2006 interagency guidance definitions, the RC-O $1 billion uninsured-deposit coverage caveat, the FDIC QBP and Risk Review anchors, and what this page deliberately is not: not SRISK, not a market-value stress test, no loan-book marking.

How FinObservatory's Bangladesh banking module is built: Bangladesh Bank source lineage, the aggregates-only design decision, definitions and coverage denominators, the IMF FSI perimeter comparison, and staleness.

How the FinObservatory credit-union layer is built: the NCUA 5300 Call Report, the whole-dollar units, the December-cycle convention for annual series and the partial-filer caveat behind it, net-worth availability from 2000Q4, and the validation anchors.

How the FinObservatory deposits layer is built: the FDIC Summary of Deposits survey and its June 30 convention, the DEPDOM reconciliation anchor, the institution-share HHI construction with the DOJ/FTC concentration thresholds, and every stated caveat.

How FinObservatory builds county banking markets from the FDIC Summary of Deposits: the branch-count HHI and the deposit HHI, the verification that the county rollup reproduces the branch file exactly, the central-booking distortion, the null-HHI case, the retired-FIPS register, and every limitation.

How FinObservatory's bank-regulator enforcement module is built: the Federal Reserve and OCC machine-readable routes, the hard individuals-exclusion rule, the conservative name-and-state cert join and its match rate, why the FDIC is absent, and coverage caveats.

How the FinObservatory FR Y-9C layer is built: the form, the pre-2006 lower-tier double count, the noncontrolling-interest identity break, the Tier 1 nulls, the reporting-threshold steps, the source cutoff, and what the panel cannot compute.

How the FinObservatory lending layer is built: the HMDA screen, the composition standardization, the string sentinels that are not zeros, the schema drift between the 2018 and 2019 files, the county FIPS validation, and what the public register cannot tell you.

How the FinObservatory mortgage layer is built: HMDA for applications and denials, FHFA NMDB for national market composition, the privacy modifications on the published files, the schema/code mappings, the weighting rules, suppression handling, and the snapshot-revision caveat.

How the FinObservatory conditions layer is built: the 8-component PCA FCI, the Basel III one-sided-HP credit-to-GDP gap, the credit impulse, the curated yield-curve and stress panel, the SLOOS credit cycle and Z.1 sectoral debt, NY Fed money-market rates (SOFR/EFFR and the FOMC target range, with the pre-2000 FRED DFF splice), the household balance sheet, and the discount-window lender-of-last-resort disclosures, with their windows, citations, the NFCI validation, and every stated limitation.

How the FinObservatory central-bank layer is built: the exact series keys for the Fed (WALCL, WSHOTSL, WSHOMCB), Eurosystem (ECB ILM), Bank of Japan (JPNASSETS), Bank of England (RPQB75A), and SNB (snbbipo T0); unit and date normalizations; the annual GDP denominators and their vintages; why the PBoC is absent and the BoE series is quarterly with a lag; build-time checks; and per-source attribution.

How the commodities module keeps World Bank Pink Sheet frequency, variant, unit and 2010 reference bases explicit, compares complete EIA calendar months without merging sources, and documents the existing EIA energy series.

How the consumer-credit module is built: the Federal Reserve G.19 volumes and terms of credit, the quarterly Charge-Off and Delinquency Rates, and the CFPB Consumer Credit Trends borrower-mix file, including the millions-to-billions conversion and TOTALSL = REVOLSL + NONREVSL identity check, the delinquency anchor verified verbatim against the Fed release (credit-card delinquency 2.92% for 2026:Q1), the CFPB score, age, income and state coverage matrix, the January 2010 = 100 basis for the two CFPB indices, how the module differs from the CFPB complaints on /consumer and the NY Fed household-debt panel on /conditions, series definitions, and every limitation.

How the corporate-credit module is built: the publication boundary that keeps the ICE BofA OAS series off the page (internal use per the FRED series notes, one cited quotation of the latest readings, links to FRED), the April 2026 trailing-window truncation facts, the Moody's Baa-Aaa and Baa-Treasury century series with the no-splice rule and the BAA10Y continuity check, percentile definitions, the SIFMA issuance decision, the Lopez-Salido-Stein-Zakrajsek framing citation, and the ICE and Moody's license notices verbatim.

How the FinObservatory derivatives layer is built: the three BIS datasets (OTC outstanding, exchange-traded, Triennial FX turnover), the notional vs gross-market-value vs gross-credit-exposure measures, net-net and net-gross bases, the 200% currency-share convention, vintage caveats, build-time verification anchors, and the BIS attribution license.

How the expectations layer uses Philadelphia Fed SPF consensus and respondent microdata for latest median paths, dispersion, forecast errors, revisions, and unemployment vintages, with a dynamic on-disk survey vintage.

How the /flows/fed-balance-sheet route is built from the Federal Reserve's H.4.1 release: the exact DDP mnemonics used for total assets, Treasuries, MBS, loans, MS Facilities 2020 LLC, central bank liquidity swaps, currency in circulation, reserve balances, reverse repo, and the Treasury General Account, plus the two derived buckets on the page.

How the FinObservatory funding layer is built: the OFR Short-Term Funding Monitor API endpoints and the 20 curated repo and money-market-fund mnemonics, transaction vs outstanding volume, the Preliminary-vintage and published-gap conventions, the build-time additivity and anchor checks, and the US-government-work license note.

How the /flows/sector-balance-sheets module is built: the fixed 20-series sector subset and the FinObservatory market-based runnable-liabilities grouping taken from the Federal Reserve's Z.1 tables.

Series inventory, units, release vintage, sparse-cell treatment, and latest-quarter reconciliation for the Federal Reserve Z.1 levels matrix.

Series inventory, units, release vintage, quarterly coverage, and latest-quarter sector identity for the Z.1 net lending and borrowing table.

Series registers, units, exclusions, release vintage, and latest-quarter reconciliations for Treasury securities, corporate equities, and mortgages in the Federal Reserve Z.1.

Series register, hierarchy, units, exclusions, release vintage, and latest-quarter reconciliations for the Federal Reserve Z.1 household balance sheet.

How the FinObservatory money-market-fund layer is built: the SEC Form N-MFP data sets, the report-month normalisation and feeder-fund exclusion, the Item A.10 category mapping across the 2024 N-MFP3 form revision, the asset-weighted WAM and WAL, the Item C.6/C.7 composition buckets, and the build-time reproduction of the SEC's own Money Market Fund Statistics report.

How FinObservatory computes the Fama-French factor returns: percent units, geometric compounding, missing-value sentinels, the two conflicting Mkt-RF series, ragged industry coverage, and what the data cannot support.

How the FinObservatory rates layer is built: Treasury's own description of the daily par yield curve (monotone convex method, bid-side quotations), the tenor-set evolution by era with the locate-by-name rule, the FiscalData auction fields, the announcement-vs-result distinction, and the units of every figure.

How FinObservatory computes real returns on equities, housing, bonds and bills from the Jorda-Schularick-Taylor Macrohistory Database: deflation, geometric averaging, the war years, and what the housing series cannot tell you.

How the FinObservatory positioning layer is built: the CFTC Commitments of Traders historical compressed files, the 15 curated market codes, TFF vs legacy trader classes in the CFTC's own words, net-share normalization and percentile definitions, the build-time additivity and anchor checks, and the public-domain license note.

How the securitization module is built: the Federal Reserve Z.1 Financial Accounts series for the agency and private-label channels, the two accounting breaks it is built around (the 2010 FAS 166/167 GSE consolidation and the post-2010 sponsor consolidation of ABS trusts), the two anchors verified verbatim against FRED (agency mortgage securitization $10,914,011 million for 2026:Q1; private-label RMBS $2,354,889 million at its 2007:Q2 peak), how it differs from the HMDA originations on /mortgage and the G.19 volumes on /consumer-credit, series definitions, and every limitation including the auto/card/student cut Z.1 does not publish.

Methods, sources and validation for LRMES, MES, CoVaR, corrected Delta-CoVaR and the absorption ratio. Dollar SRISK remains unavailable.

How the FinObservatory valuation layer is built: source files and vintages, the CAPE definition and its reproduction check, Damodaran equity and country risk premium selections, industry-table identities, and KPSS patent-value aggregation rules.

Where the China lending figures come from, how they are constructed, what they exclude, and why they are not comparable with borrower-reported external debt.

The exact BIS selection behind every figure on /claims: one measure, one basis, aggregates excluded, and the traps in the locational banking statistics that this layer is built to avoid.

How the FinObservatory cross-border layer is built: BIS global liquidity, locational and consolidated banking statistics, plus the Treasury TIC holder-side view on /crossborder/USA, with exact selections, units and coverage caveats.

How FinObservatory builds external balance-sheet and bilateral direct-investment pages: source lineage, reported and counterpart-derived positions, the equity-versus-debt split, and data limits.

How the FinObservatory external-debt layer is built: the World Bank Debtor Reporting System, the 24 IDS indicators and their exact definitions, the derivation of the bilateral residual, the cut between actual years and the contracted debt-service schedule, the balanced panel, and the coverage caveats.

How the FinObservatory state-and-local public-finance layer is built: the Census Annual Survey of State and Local Government Finances and the Federal Reserve Z.1 municipal-debt series, their units and definitions, the vintage, and every stated caveat.

How the FinObservatory remittances layer is built: the WDI balance-of-payments series, the GDP denominator, the balanced panel, the page rule, and why the bilateral corridor matrix is a model output rather than a measurement.

How the FinObservatory reserves layer is built: the IMF COFER dataset and dimensions, the World-aggregate allocated-reserve shares, the five-currency subset limitation, the recent-quarter revision caveat, and the exact queries behind every figure.

How the FinObservatory sovereign-debt layer is built: the IMF debt series and their fallback chain, the debt-at-crisis construction, the era distributions, the current-elevation benchmark, and every stated caveat.

How the FinObservatory analyst works: the never-generate-numbers rule, the StatGPT architecture, the fixed tool set, the citation format, what it will not do, and the model and privacy disclosures.

How the FinObservatory country pages are composed: the union of layers that defines which countries get a page, the exact GFDD and BIS series shown, each source's coverage and staleness, and the license the composition inherits.

How the FinObservatory consumer-complaint layer is built: the CFPB Consumer Complaint Database bulk export pipeline, the no-narrative no-PII aggregation rule, the product-family mapping across the form's label changes, the Census-via-FRED per-capita denominator, the partial-period cut, the bulk-vs-API cross-checks, and the Bureau's own publication criteria and caveats quoted verbatim.

How the /filings module is constructed from the SEC Financial Statement Data Sets: the universe, the curated tag lists, the five conditions that make a fact safe to sum, and what the module refuses to claim.

How FinObservatory's financial-crime reference layer is built: the two FATF lists quoted verbatim, why the R.1-R.40 / IO.1-IO.11 ratings are not reproduced, the OFAC aggregates-only decision, vintages, and license notes.

How the FinObservatory inclusion layer is built: the Findex survey rounds and how 2022 folds into the 2021 round, the indicators published, the coverage breaks, why GFDD is not mixed with Findex, and every stated caveat.

Sources, licenses, units and the estimate contests behind the inequality module: WID.world and Auten-Splinter top shares, World Bank PIP Ginis and 2021-PPP poverty lines, OECD disposable-income Ginis, and the UNU-WIDER WIID consistency layer. Why income definitions matter, why fiscal-data top shares and survey Ginis are different objects, and why the poverty-line PPP vintage is pinned.

How the FinObservatory research library was acquired and what it may show: per-corpus acquisition detail, the recorded license language for BIS, Federal Reserve, IMF, and SEC material, the catalog schema, the federalreserve.gov URL mapping, and the honest-absence rule.

How the FinObservatory mobility layer is built: absolute versus relative mobility, income versus education; the Opportunity Insights fading-American-dream cohort series and the Opportunity Atlas upward-mobility measure with its published standard errors; the World Bank GDIM education-persistence measures and parent definitions; the in-house Great Gatsby curve with sources stated per axis; the generation lag that ends every series a generation ago; Bangladesh coverage; build-time verification anchors; and the Opportunity Insights and World Bank license notes.

How the FinObservatory payments layer is built: the BIS CPMI bulk datasets and the canonical instrument-total cuts, the Fedwire Funds annual and monthly tables, the CHIPS-via-CPMI route and its TCH cross-check, the WDI population and GDP denominators with vintages, the build-time additivity and cross-source anchors, and the attribution-only license posture.

How the FinObservatory redistribution layer is built: the income concepts (market, gross, disposable; CBO's before and after transfers and taxes), the market-minus-disposable Gini measure of fiscal redistribution, the OECD IDD 2012 methodology break, the gross-versus-net social-spending nuance, the COFOG concept boundary, what fiscal redistribution excludes (predistribution and in-kind services), build-time verification anchors, and the OECD, IMF and CBO license notes.

How FinObservatory reads the Federal Reserve's Distributional Financial Accounts and Survey of Consumer Finances: the DFA construction, revision and unit caveats, why surveys, capitalization and heterogeneous-returns methods disagree about top wealth, and why there is no DFA median.