BIS Locational Banking Statistics (LBS) & BIS Data Portal

The official map of who owes whom across borders. The Bank for International Settlements compiles the locational banking statistics under the Committee on the Global Financial System, measuring international banking from a residence perspective - what the banking offices located in each of about 50 reporting countries claim from, and owe to, residents of more than 200 counterparty countries, quarter after quarter. The current release carries roughly 609,000 series rows x 225 columns: 24 dimension/metadata columns followed by one column per quarter from 1977-Q4 through 2026-Q1.

Three things make this table hard to substitute. First, reach: the BIS estimates the reporting perimeter captures about 95% of all cross-border banking activity - near-census coverage no commercial panel assembles. Second, the axis set: every series is keyed by counterparty sector (banks versus non-banks), instrument (loans and deposits, debt securities, currency and deposits), currency denomination and position type (cross-border versus local positions of foreign-owned offices). Third, the derived measures: alongside raw stocks, FX- and break-adjusted changes and annual growth rates ride in the same release, computed upstream so valuation swings do not impersonate new lending.

Within Datadory's diversified-banks shelf this is the only dataset that sees the whole world's cross-border bank balance sheet in one schema. US regulators' filings stop at the border; this one starts there. Get a sample of this dataset cut to your counterparties and measures, and the first rows arrive with your filter already applied.

Each row is one fully-keyed series; the quarterly values run out to its right. Two real rows, exactly as they arrive:

freq          : Q
measure       : S  (amounts outstanding)
instrument    : G  (loans and deposits)
currency      : JPY (yen)
reporting cty : 5A (all reporting countries)
cp sector     : N  (non-banks, total)
cp country    : YE (Yemen)
position      : N  (cross-border)
collection    : E  (end of period)
quarters      : 1977-Q4 ... 2026-Q1   # one value column per quarter

freq          : Q
measure       : F  (FX and break adjusted change)
position      : C  (total claims)
reporting cty : BR (Brazil)
currency type : F  (foreign currency)
position type : R  (local)
collection    : S  (summed through period)
1977-Q4 value : 1643.983

Swap the counterparty code and the same key returns China, Switzerland or any offshore centre; swap measure and stocks become flows with the valuation noise removed. Name your reporting countries, counterparties and measures when you request a sample and Datadory will cut exactly that slice.

Eleven documented fields key the panel, verified against the released file during the research pass - nothing inferred from column names alone. Every series is the cross product of these dimensions, which is why the same quarter can be sliced by debtor country, by sector, by currency or by all three at once without a re-pull.

Series also carry a visibility flag and period-format descriptor, and some deliveries add derived columns such as rolling multi-quarter aggregates, custom country groupings or aligned extracts joined to the companion consolidated statistics. These fold under additional fields on request - their shape depends on the cut you specify, so they are confirmed and populated when you ask for a sample.

Geography - about 50 reporting countries across Europe, the Americas, Asia-Pacific and the offshore centres, against counterparties in more than 200 countries. Aggregates (5A all reporting countries, 5C euro area, 5J all counterparty countries) sit beside the individual codes, so global totals are a filter rather than a merge.

Temporal - quarterly observations from 1977-Q4 through 2026-Q1 in the current release, 195 quarters per series. Long enough to hold Latin America's lost decade, the Asian crisis, the euro area's sovereign stress and the post-2022 retrenchment of offshore lending inside one continuous panel.

Granularity - reporting country x counterparty country x counterparty sector x instrument x currency x position type x measure, per quarter, in millions of US dollars (or growth ratios where the measure calls for them).

Set against the wider catalog - average quality score 7.81 across all 1,744 datasets - this slice scores 10/10, carried by complete field documentation and a time series no private vendor reconstructs from scratch.

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

• Global-liquidity modeling - dollar- and yen-denominated cross-border credit to non-banks is the canonical gauge of the global financial cycle; measure F strips the valuation noise that makes raw stocks lie about momentum.
• Funding-stress and carry backtests - quarterly claims and liabilities from 1977-Q4 onward give strategies five decades of regime changes to prove themselves against, not just the last cycle.
• Offshore-centre exposure mapping - position type R isolates local lending by foreign-owned offices, the book most risk registers miss when they count only cross-border claims.
• Country-risk screening - counterparty-level claims by sector let credit committees size what a given sovereign's banking system actually owes abroad before the headline number gets politicized.
• Policy and journalism - the series names its compiler, method and residence basis, which is why central banks, academics and financial journalists cite it directly.

Investors and quant researchers backtest funding-stress and carry signals on quarterly cross-border flows spanning five decades - the depth is the point. Data scientists and ML engineers model global liquidity on a panel whose schema has held since the 1970s, so features engineered today survive the next release. Developers building data products wire the dimension tuple into risk dashboards redrawn on whatever cadence the product needs. Journalists, academics and students ground stories on cross-border and offshore lending in the official record rather than a secondary chart. Industry context sits on the diversified banks data hub; persona workflows at investors & quants x diversified banks, data scientists x diversified banks, developers & builders x diversified banks and journalists & academics x diversified banks.

Against the rest of the shelf: the FDIC Quarterly Banking Profile aggregates the US industry's earnings and asset quality - a domestic income statement, not a cross-border balance sheet. ECB Consolidated Banking Data covers euro-area banks' capital and funding in real depth but stops at the EU perimeter. SEC EDGAR XBRL resolves individual listed banks line by line, one issuer at a time. Only this table answers which countries' banks owe how much to which other countries, in which currency - and its natural companion, the BIS consolidated statistics, recuts the same activity by bank nationality when ultimate-risk attribution matters more than location. The head-to-head against the FDIC product is worked through in BIS LBS vs FDIC QBP.

Four honest caveats. First, aggregation level: rows describe a reporting country's banking sector in aggregate - no individual bank is named, so bank-level questions want filings data joined alongside. Second, breaks happen: when reporters revise methods or enter or leave the perimeter, the B break measure marks the discontinuity, and long-run panels should respect it rather than splice blindly. Third, currency composition: positions are converted into US dollars, so raw period-to-period moves mix genuine flows with valuation effects - use the FX- and break-adjusted change measure when direction matters more than level. Fourth, cadence: the panel moves quarterly, so intra-quarter events surface one release later than they would in market data. None of these bite unexpectedly; they ship flagged against the analysis you plan to run.

Because the raw artifact is a 537 MB wide file built for archivists, and most questions want one slice. Datadory cuts samples to your reporting countries, counterparties, currencies and measures, schedules deliveries into your warehouse keyed on the dimension tuple so consecutive quarters diff cleanly, and flags the breaks before they corrupt a backtest. Browse the rest of the industry on the diversified banks data hub, the best diversified-banks datasets ranking, or the full catalog.