BIS Debt Securities and Capital Markets Statistics

The official answer to who owes what in the world's bond markets. BIS Debt Securities and Capital Markets Statistics - filed under Diversified Capital Markets, compiled by the Bank for International Settlements - records how much debt securities residents of more than 50 economies have issued and how much remains outstanding, cut by issuing sector, instrument type, original and residual maturity band, currency denomination and valuation basis.

Sector reaches from central government down to households; instruments span treasury bills, commercial paper, negotiable certificates of deposit, bonds and debentures, asset-backed securities, and green and sustainable instruments. Because definitions follow the Handbook on Securities Statistics, a rupiah series and a rand series read identically - domestic versus international issuance turns on where the security was issued relative to the borrower's residence, not on anyone's national habit.

Quarterly and annual observations run back as early as 1946, and government bond series cover 54 countries - 28 advanced, 26 emerging - beside euro area and world aggregates. Get a sample of this dataset and we will return real series rows with the full dictionary attached.

Two observations from the same quarter, exactly as they ship:

REF_AREA      : ID      Indonesia
REF_SECTOR    : S1311   central government
STO           : LE      closing balance-sheet stocks
INSTR_ASSET   : F3      debt securities
MATURITY      : L       original maturity above 1 year
UNIT_MEASURE  : IDR     Indonesian rupiah
UNIT_MULT     : 9       values in millions
TITLE         : Indonesia - Debt sec, issued by central gov,
                all markets, lt org mat (> 1y or no stated maturity),
                all currencies, nominal value, stocks
TIME_PERIOD   : 2025-Q4
OBS_VALUE     : 8234861.4676094

REF_AREA        : ZA      South Africa
REF_SECTOR      : S1      total economy
STO             : LE      closing balance-sheet stocks
INSTR_ASSET     : F3      debt securities
CUST_BREAKDOWN  : L_SX    sustainability-linked, second party opinion
TIME_PERIOD     : 2025-Q4
OBS_VALUE       : 55.74840514609

Read them together and the design falls out. The Indonesia row sizes the whole longer-term central-government stock in millions of rupiah, UNIT_MULT doing the scaling so OBS_VALUE stays a bare number. The South Africa row ignores maturity and currency splits entirely and instead switches on the custom-breakdown axis to isolate sustainability-linked instruments by second-party opinion - the green-finance question riding the same grid as the plain-vanilla totals.

Notice also which keys each row carries. A series splits along a dimension or it does not, and the absent key tells you which - no placeholder codes, no sentinel values to strip.

Fifteen fields define every observation, all verified against live records rather than documentation alone. Thirteen are selectors - frequency, reference area, counterpart area, issuing sector, accounting entry, measure type, instrument, maturity, unit, multiplier, valuation, custom breakdown - and the last two carry the readable title and the number itself. Nothing else is inferred:

Geography - over 50 economies for debt securities statistics, with government bond series stretching to 54 countries split 28 advanced and 26 emerging. Euro area and world aggregates ride beside the country rows, so a global chart and a single-market view come from one schema.

Temporal - quarterly and annual observations beginning as early as 1946; our verified sample rows run through 2025-Q4. Eight decades of depth is the difference between a trend line and an anecdote.

Granularity - one observation per fully keyed combination of issuing sector, instrument, maturity band, currency and valuation, at quarterly frequency. The single-topic table spans roughly 469 columns and hundreds of thousands of series cells, which is what a fifty-economy panel times eighty years actually looks like.

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

You pick the channel and the cadence; parsing dimension codes, resolving labels and holding the schema steady are our problem. When a new quarter lands or an economy's series is restated, the feed you consume stays normalized - same columns, same types, same join keys. Tell us the cut you need - one economy, one sector, or the full 1946-to-present panel - and the sample arrives shaped that way: Get a sample of this dataset, rows and dictionary included.

• Fixed-income research - issuance and outstanding amounts by sector, maturity and currency give sell-side and buy-side desks the supply side of bond pricing, quarter after quarter, without stitching national sources.
• Cross-country backtesting - harmonised panels reaching back as far as 1946 let quant teams test bond-market strategies across 50-plus economies on one definition instead of eighty incompatible ones.
• Market-depth comparison - capital-market depth across 50-plus economies measured on Handbook-on-Securities-Statistics definitions makes the 'how developed is this bond market' slide defensible.
• Green-finance tracking - sustainability-linked instrument splits with second-party-opinion assurance levels turn ESG issuance claims into counted series.
• Macro and policy monitoring - central-government and household sector rows quantify who is leveraging up, the input behind debt-sustainability and financial-stability work.
• Citation-grade journalism and research - one citable official series per claim about global debt markets, attributed to the BIS rather than a press release.

Investors and quants (3/3 relevance) get the supply-side series behind fixed-income models - issuance and outstanding by sector, maturity and currency, deep enough to backtest on. See investors quants use cases.

Data scientists and ML engineers (3/3) get quarterly debt-securities series shaped as ready-made cross-country panels, with a stable typed schema that drops straight into feature pipelines. See data scientists use cases.

Market researchers and consultants (2/3) get bond-market sizing across 50-plus economies on one methodology - the benchmark layer for capital-markets briefs. See market researchers use cases.

Developers and builders (2/3) get a stable dimensional schema where every column is a documented selector, trivially mapped into filters and dashboards. See developers builders use cases.

Journalists, academics and students (2/3) get the citable record behind global bond-market claims. Competitive-intel and product teams (1/3) get bond-market growth curves for the economies where capital-markets clients compete. See competitive intel product teams use cases.

Three things worth having upfront. First, valuation basis changes the story: market value and nominal value coexist inside the same dataset, so fix VALUATION before comparing economies - a market-value total moves with prices, a nominal stock does not.

Second, values arrive scaled: UNIT_MULT carries the power of ten (9 means millions), so naive sums mix magnitudes unless the multiplier is applied first. We resolve units into labeled columns as part of the delivery, which removes the classic wrong-by-a-thousand error.

Third, aggregates sit beside countries: euro area (U2) and world rows share the reference-area field with sovereign states, so filter aggregates out before averaging or you will double-count continents. Say which economies, sectors and periods you need when you request a sample - the extract comes back cut to that shape.

Notes worth reading next:

• Diversified capital markets data hub - the pooled view of the industry slice, nine primary datasets deep, from regulator transparency to exchange statistics.
• BIS Full Data Sets Bulk Downloads - the breadth counterpart: all 27 BIS topics in one shared archive grammar when debt securities is only part of the job.
• Debt securities statistics vs BIS Full Data Sets - single-topic depth against twenty-seven-topic breadth, scored honestly.
• Issuance and outstanding amounts, explained - the two measures this dataset exists to quantify, unpacked.
• Best diversified capital markets datasets - where this record ranks within the slice and what beats it for other jobs.