OTC derivatives data
Liquidity and pricing in the over-the-counter (OTC) derivative markets have tended to have less transparent than their exchange traded (lit) equivalents. The Bank of International Settlements (BIS) has published statistics of trading in the OTC markets on a triennial basis. However, representative statistics of volume and pricing data on a more frequent basis has been near impossible to obtain. Post-GFC derivatives reforms have lifted the veil in these markets.
Swap execution facilities
Some Over-The-Counter (OTC) derivatives have been Made Available for Trade (MAT) by the CTFC. This means they must be traded on Swap Execution Facilities (SEFs). These SEFs are compelled publish trading volume and prices for these MAT derivatives on a daily basis.
Some of the most widely used SEFs are run by Bloomberg, ICAP, Tullett Prebon and Tradeweb among others.
Swap data repositories
The key economic terms of traded swaps must be reported to an authorised Swap Data Repository (SDR). Some of the most widely used SDRs are the DTCC Data Repository, ICE Trade Vault and the CME’s SDR. The CFTC provides weekly snapshots of data collected by these SDRs on a weekly basis. SDRs domiciled in different regulatory jurisdictions are expected to provide differing levels of data. U.S. regulations compel U.S. domiciled SDRs to provide (anonymised) trade level data to the public while SDRs in other jurisdictions (e.g. in Europe) expected far less granular, and typically only aggregated, data.
dataonderivatives
You can source OTC derivatives data from supported sources. Supported sources include BloombergSEF and several U.S. domiciled SDRs including DTCC’s DDR, Bloomberg’s SDR and CME. SDRs in other jurisdictions do not provide trade level data and consequently these sources are unsupported. Further sources that provide trade level data will be added over time (and pull requests to that end are welcome).
Usage
You can download daily trade data executed on the BloombergSEF:
library("dataonderivatives")
bsef(as.Date("2021-12-13"), asset_class = "IR")
#> # A tibble: 333 × 13
#> tradeDate security priceOpen priceHigh priceLow priceClose settlementPrice
#> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
#> 1 2021-12-13… AUD SWAP… 2.234 2.234 2.234 2.234 2.234
#> 2 2021-12-13… AUD SWAP… 2.55625 2.55625 2.26 2.26 2.26
#> 3 2021-12-13… CAD SWAP… 1.94 1.94 1.94 1.94 1.94
#> 4 2021-12-13… CAD SWAP… 1.363362 1.4839 0.87616 0.87616 0.87616
#> 5 2021-12-13… CAD SWAP… 2.419 2.419 2.419 2.419 2.419
#> 6 2021-12-13… CAD SWAP… 2.419 2.419 2.419 2.419 2.419
#> 7 2021-12-13… CAD SWAP… 2.3765 2.3765 2.3585 2.3585 2.3585
#> 8 2021-12-13… CHF SWAP… -0.4344 -0.4344 -0.4344 -0.4344 -0.4344
#> 9 2021-12-13… CHF SWAP… -0.25 -0.25 -0.25 -0.25 -0.25
#> 10 2021-12-13… EUR MAC … 0.25 0.25 0.25 0.25 0.25
#> # … with 323 more rows, and 6 more variables: totalVolume <chr>,
#> # blockTradeVolume <chr>, totalVolumeUsd <chr>, blockTradeVolumeUsd <chr>,
#> # currency <chr>, premiumVolatilityOtherFlag <chr>
# Get more than one day's data
bsef(as.Date("2021-12-13"), as.Date("2021-12-17"), asset_class = "IR")
#> # A tibble: 1,523 × 13
#> tradeDate security priceOpen priceHigh priceLow priceClose settlementPrice
#> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
#> 1 2021-12-13… AUD SWAP… 2.234 2.234 2.234 2.234 2.234
#> 2 2021-12-13… AUD SWAP… 2.55625 2.55625 2.26 2.26 2.26
#> 3 2021-12-13… CAD SWAP… 1.94 1.94 1.94 1.94 1.94
#> 4 2021-12-13… CAD SWAP… 1.363362 1.4839 0.87616 0.87616 0.87616
#> 5 2021-12-13… CAD SWAP… 2.419 2.419 2.419 2.419 2.419
#> 6 2021-12-13… CAD SWAP… 2.419 2.419 2.419 2.419 2.419
#> 7 2021-12-13… CAD SWAP… 2.3765 2.3765 2.3585 2.3585 2.3585
#> 8 2021-12-13… CHF SWAP… -0.4344 -0.4344 -0.4344 -0.4344 -0.4344
#> 9 2021-12-13… CHF SWAP… -0.25 -0.25 -0.25 -0.25 -0.25
#> 10 2021-12-13… EUR MAC … 0.25 0.25 0.25 0.25 0.25
#> # … with 1,513 more rows, and 6 more variables: totalVolume <chr>,
#> # blockTradeVolume <chr>, totalVolumeUsd <chr>, blockTradeVolumeUsd <chr>,
#> # currency <chr>, premiumVolatilityOtherFlag <chr>
You can also download the data reported to the DTCC’s and CME’s SDRs:
ddr(as.Date("2021-12-13"), "IR")
#> # A tibble: 13,891 × 69
#> `Dissemination ID` `Original Dissemi… `Primary Asset C… `Product ID` Action
#> <dbl> <dbl> <chr> <chr> <chr>
#> 1 248871220 NA IR InterestRate:… NEW
#> 2 248871315 NA IR InterestRate:… NEW
#> 3 248871316 NA IR InterestRate:… NEW
#> 4 248871317 NA IR InterestRate:… NEW
#> 5 248871481 248870394 IR InterestRate:… CANCEL
#> 6 248871482 248870394 IR InterestRate:… CORRE…
#> 7 248771101 NA IR InterestRate:… NEW
#> 8 248770933 NA IR InterestRate:… NEW
#> 9 248815618 248815527 IR InterestRate:… CANCEL
#> 10 248815621 248815582 IR InterestRate:… CANCEL
#> # … with 13,881 more rows, and 64 more variables: Transaction Type <chr>,
#> # Block Trade Election Indicator <chr>, Cleared <chr>,
#> # Clearing Exception or Exemption Indicator <chr>,
#> # Day Count Convention <chr>, Effective Date <date>,
#> # Embedded Option Type <chr>, Event Timestamp <dttm>, Exchange Rate <lgl>,
#> # Exchange Rate Basis <lgl>, Execution Timestamp <dttm>,
#> # Expiration Date <date>, First Exercise Date <date>, Fixed Rate 1 <dbl>, …
cme(as.Date("2021-12-13"), "CO")
#> # A tibble: 406 × 64
#> Event `Execution Timestamp` `Dissemination Time` Cleared Collateralizati…
#> <chr> <chr> <chr> <chr> <chr>
#> 1 New Trade 12/10/2021 14:36:48 12/10/2021 14:36:48 Unclea… Partially Colla…
#> 2 New Trade 12/10/2021 14:44:23 12/10/2021 14:44:23 Unclea… Partially Colla…
#> 3 New Trade 12/10/2021 16:35:19 12/10/2021 16:35:19 Unclea… Partially Colla…
#> 4 New Trade 12/10/2021 14:44:54 12/10/2021 14:44:54 Unclea… Partially Colla…
#> 5 New Trade 12/10/2021 14:44:54 12/10/2021 14:44:54 Unclea… Partially Colla…
#> 6 New Trade 12/10/2021 23:26:03 12/10/2021 23:26:03 Unclea… Partially Colla…
#> 7 New Trade 12/10/2021 15:57:47 12/10/2021 15:57:47 Unclea… Partially Colla…
#> 8 New Trade 12/10/2021 17:04:18 12/10/2021 17:04:18 Unclea… Partially Colla…
#> 9 New Trade 12/10/2021 18:41:16 12/10/2021 18:41:16 Unclea… Partially Colla…
#> 10 New Trade 12/10/2021 22:19:05 12/10/2021 22:19:05 Unclea… Partially Colla…
#> # … with 396 more rows, and 59 more variables: End-User Exception <lgl>,
#> # Bespoke <chr>, Block/Off Facility <chr>, Execution Venue <chr>, UPI <lgl>,
#> # Product <chr>, Asset Class <chr>, Contract Type <chr>,
#> # Effective Date <chr>, Maturity Date <chr>, Sub Asset Class <chr>,
#> # Leg 1 Total Notional <lgl>, Price <dbl>, Settlement Method <chr>,
#> # Settlement Currency <chr>, Leg 1 Type <chr>, Leg 1 Fixed Payment <dbl>,
#> # Leg 1 Fixed Payment Currency <chr>, Leg 1 Index <chr>, …