- Reference
- 2026-09-08
- Actual
- 3.63
- Previous
- 3.64
- Forecast
- --
United States Tri-Party General Collateral Rate
The Tri-Party General Collateral Rate series for United States is a key macroeconomic indicator published daily by the national statistics authority or the Federal Reserve. It is available via the FXMacroData API at /api/v1/announcements/usd/tgcr.
- Latest
- 3.63
- Release date
- 2026-09-09
- Previous
- 3.64
- Change
- -0.01
- History starts
- Updating
Reference: 2026-09-08
Release alerts
Get USD Tri-Party General Collateral Rate alerts
Enter your account email. We will notify you when the next official-source USD Tri-Party General Collateral Rate release is published.
Why Tri-Party General Collateral Rate matters for USD
United States' Tri-Party General Collateral Rate is monitored by the Federal Reserve when assessing the health of the economy and calibrating monetary policy. Surprises relative to consensus can move the usd in the short term.
How to interpret this series
A reading above market expectations typically signals stronger-than-anticipated conditions and is broadly supportive of the usd. A miss versus consensus can shift rate expectations and weigh on the usd.
What this data measures
Tri-Party General Collateral Rate is a published macroeconomic indicator covering one or more of the currencies supported by FXMacroData. The New York Fed measure of overnight tri-party general collateral repo borrowing costs secured by U.S. Treasury securities, expressed as an annual percentage rate. The table on this page shows the latest released value for every supported currency, the previous reading, the change between the two, the reference date, the publication frequency, the units, and the original publishing source.
How to read the release
Compare across currencies and watch the spread to the policy rate target. Persistent spreads can signal funding stress or liquidity imbalances that move the currency on the margin.
Historical Tri-Party General Collateral Rate
Source: Federal Reserve Bank of New York. Cadence: Daily. Unit: %. Coverage metadata updating.
Recent announcements
Latest release rows with direct links to the announcement detail pages.
| Release time | Reference period | Actual | Previous | Forecast | Detail |
|---|---|---|---|---|---|
|
2026-09-09 12:00 UTC
|
3.63 | 3.64 | -- | Announcement page | |
|
2026-09-08 12:00 UTC
|
3.64 | 3.64 | -- | Announcement page | |
|
2026-09-04 12:00 UTC
|
3.64 | 3.63 | -- | Announcement page | |
|
2026-09-03 12:00 UTC
|
3.63 | 3.63 | -- | Announcement page | |
|
2026-09-02 12:00 UTC
|
3.63 | 3.67 | -- | Announcement page | |
|
2026-09-01 12:00 UTC
|
3.67 | 3.63 | -- | Announcement page | |
|
2026-08-31 12:00 UTC
|
3.63 | 3.62 | -- | Announcement page | |
|
2026-08-28 12:00 UTC
|
3.62 | 3.62 | -- | Announcement page | |
|
2026-08-27 12:00 UTC
|
3.62 | 3.64 | -- | Announcement page | |
|
2026-08-26 12:00 UTC
|
3.64 | 3.63 | -- | Announcement page |
- Reference
- 2026-09-04
- Actual
- 3.64
- Previous
- 3.64
- Forecast
- --
- Reference
- 2026-09-03
- Actual
- 3.64
- Previous
- 3.63
- Forecast
- --
- Reference
- 2026-09-02
- Actual
- 3.63
- Previous
- 3.63
- Forecast
- --
- Reference
- 2026-09-01
- Actual
- 3.63
- Previous
- 3.67
- Forecast
- --
- Reference
- 2026-08-31
- Actual
- 3.67
- Previous
- 3.63
- Forecast
- --
- Reference
- 2026-08-28
- Actual
- 3.63
- Previous
- 3.62
- Forecast
- --
- Reference
- 2026-08-27
- Actual
- 3.62
- Previous
- 3.62
- Forecast
- --
- Reference
- 2026-08-26
- Actual
- 3.62
- Previous
- 3.64
- Forecast
- --
- Reference
- 2026-08-25
- Actual
- 3.64
- Previous
- 3.63
- Forecast
- --
United States Tri-Party General Collateral Rate API docs
Developer reference, data provenance, response schema, and examples are kept on this same country indicator page so the public research page and API documentation do not compete with each other.
Endpoint coverage and contract
This page documents the exact request inputs, response fields, access model, and currently published history window for
/api/v1/announcements/usd/tgcr.
Coverage metadata updating
Production OpenAPI schema: https://api.fxmacrodata.com/openapi.json
Official series ID: NYFed:reference-rates:TGCR:percentRate
USD is free without a key. All other currencies require an Individual API key.
Public USD endpoint — anonymous callers receive the most recent 90 days. Add `?api_key=YOUR_API_KEY` to page through full stored USD history or to access any other currency. Date filters are applied before pagination. The default page size is 20 rows and the maximum is 100. Use either `page` or `offset`; when both are provided, `page` takes precedence. Broad windows should be traversed with `pagination.next_offset`, not split into custom shorter-window retries. If the selected series exists but a requested date window has no observations, the API returns 200 with an empty `data` array and a `no_data` object (`error_code` `NO_DATA_IN_REQUESTED_WINDOW`) carrying coverage dates, `recommended_start_date`, and `available_observation_count`; a 404 means the currency or indicator slug is wrong. Forecasts are served separately by /v1/predictions/usd/tgcr; join via `announcement_id`.
Macroeconomic announcement and catalogue rows are normalized from official central-bank, national statistics, treasury, or government publisher releases. Responses expose publisher metadata via source, source_url, and provenance fields where available.
Responses expose source, source_url, provenance, and data_quality fields for source citation, timestamp completeness, stale-data flags, and point-in-time safety checks.
Optional indicators use per-currency fetcher capability checks and may be temporarily hidden while source coverage is being remediated.
foreign_reservescb_assetsinflation_momPath parameters
| Parameter | Required | Type | Description |
|---|---|---|---|
currency |
YES | string |
3-letter currency code. |
indicator |
YES | string |
Indicator slug. Use /v1/data_catalogue/{currency} to list available slugs per currency. |
Query parameters
| Parameter | Required | Format | Description |
|---|---|---|---|
start_date |
NO | string (YYYY-MM-DD) |
Earliest observation date to include. Date filters are applied before pagination. Authenticated requests page through full stored history when omitted; anonymous USD requests are limited to the most recent 90 days and 100 no-key requests/day. |
end_date |
NO | string (YYYY-MM-DD) |
Latest observation date to include. Defaults to today. |
limit |
NO | integer |
Maximum number of most-recent-first rows to return per page. Defaults to 20; maximum 100. |
page |
NO | integer |
One-based page number. When supplied, page takes precedence over offset; page=2 with limit=20 is equivalent to offset=20. |
offset |
NO | integer |
Zero-based row offset after date filtering. Use with pagination.next_offset to request the next page. |
api_key |
Individual (USD: no-key 100/day) | string |
Individual API key. Required for non-USD currencies and for USD requests that need history older than 90 days. |
Use start_date and end_date to narrow the series first, then page through the filtered result with limit plus either page or offset.
The default page size is 20 rows and the maximum is 100.
The response includes earliest_available_date, latest_available_date, and a pagination object with returned_count, total_count, has_more, next_offset, and page_includes_latest_available.
If the selected series exists but the requested date window has no rows, the endpoint returns 200 with an empty data array and a no_data object (error_code=NO_DATA_IN_REQUESTED_WINDOW) naming the coverage the series does have.
Response fields
| Field | Type | Description |
|---|---|---|
currency |
string |
3-letter currency code. |
indicator |
string |
Indicator slug as requested. |
has_official_forecast |
boolean |
True if the central bank publishes an official forecast for this indicator. |
start_date |
string |
Earliest date in the filtered result set (YYYY-MM-DD). |
end_date |
string |
Latest date in the filtered result set (YYYY-MM-DD). |
earliest_available_date |
string | null |
Earliest date available for this series before any date filter or pagination is applied. |
latest_available_date |
string | null |
Latest date available for this series before pagination is applied. |
requested_start_date |
string |
Start date applied to the request after endpoint defaults are resolved. |
requested_end_date |
string |
End date applied to the request after endpoint defaults are resolved. |
requested_window_has_data |
boolean |
False when the endpoint returned the latest eligible observation before the requested window instead of an observation inside the requested date range. |
page_includes_latest_available |
boolean |
True when the current page includes latest_available_date. False usually means the latest row is on a different page or the response is a historical page. |
pagination.limit |
integer |
Page size used for this response. |
pagination.offset |
integer |
Zero-based offset used for this response. |
pagination.returned_count |
integer |
Number of rows included in data[]. |
pagination.total_count |
integer |
Total number of rows matching the path and date filters before pagination. |
pagination.has_more |
boolean |
True when another page is available. |
pagination.next_offset |
integer | null |
Offset to use for the next page, or null when there are no more rows. |
pagination.page_includes_latest_available |
boolean |
Mirrors page_includes_latest_available inside the pagination envelope for clients that keep pagination state together. |
cb_target |
object | null |
Central bank target metadata (e.g. inflation target range), if applicable. |
data[].date |
string |
Observation date (YYYY-MM-DD). |
data[].announcement_id |
string |
Stable announcement identifier in the form `{currency}_{indicator}_{date}` — use it to join predictions from /v1/predictions/{currency}/{indicator} and revision history. |
data[].val |
number | null |
Observed value in the indicator's native unit. |
data[].announcement_datetime |
integer | null |
Unix timestamp (UTC) of the official data release. |
data[].release_time_assumed |
boolean | null |
Present and true when announcement_datetime was derived from the median publication lag measured on this series' confirmed releases, because no official release calendar reaches that period. Omitted on rows whose publication time was captured from the release. |
data[].pct_change |
number | null |
Period-over-period percentage change. |
data[].pct_change_12m |
number | null |
12-month rolling percentage change. |
data_quality |
object |
Standardized endpoint-level quality, source, freshness, and point-in-time safety metadata for the returned payload. |
data_quality.is_official |
boolean |
True when the response is backed by an official source and is not proxy, fallback, derived, or prohibited-source data. |
data_quality.is_proxy |
boolean |
True when the response is an explicit proxy for the requested concept rather than the primary official series. |
data_quality.is_fallback |
boolean |
True when one or more returned rows are marked as fallback data. |
data_quality.is_stale |
boolean |
True when the newest available observation is older than the endpoint's frequency-aware stale_after_days threshold. |
data_quality.has_announcement_datetime |
boolean |
True when all rows in the quality scope carry the required announcement timestamp field or fields. |
data_quality.point_in_time_safe |
boolean |
True only when every row in the quality scope carries a publication timestamp captured from the release itself, and the result is not fallback or prohibited-source data. One row with a missing or derived timestamp clears it for the whole response, so most full-history requests return false; gate a backtest on the per-row release_time_assumed flag. |
data_quality.has_assumed_release_times |
boolean |
True when at least one row in scope carries a publication timestamp derived from the series' measured publication lag rather than captured from the release. |
data_quality.assumed_release_time_count |
integer |
Number of rows in scope whose publication timestamp was derived. Compare against row_count to size the derived portion of the window. |
data_quality.latest_available_date |
string | null |
Freshest observation date available for the quality scope before pagination-specific caveats. |
data_quality.last_updated |
string | null |
Server-side refresh or update timestamp when the endpoint has one available. |
data_quality.data_lag_days |
integer | null |
Calendar-day lag between today and latest_available_date; null when the latest date is missing, invalid, or future-dated. |
data_quality.source_name |
string | null |
Human-readable source or publisher label used for the response. |
data_quality.source_type |
official | public | fallback | derived |
Normalized source classification shared across source-backed REST and MCP responses. |
data_quality.is_derived |
boolean |
True when the endpoint output is calculated from underlying source rows, such as spreads, slopes, forward curves, forecasts, or sentiment composites. |
data_quality.row_count |
integer |
Number of rows considered in the data-quality scope. |
data_quality.announcement_datetime_count |
integer |
Number of rows in scope with complete required announcement timestamp fields. |
data_quality.missing_announcement_datetime_count |
integer |
Number of rows in scope missing at least one required announcement timestamp field. |
data_quality.quality_scope |
string |
Scope used to compute the quality object, such as full_result, latest, recent, or scheduled_releases. |
data_quality.stale_after_days |
integer | null |
Frequency-aware threshold used for is_stale; daily=7, weekly=21, monthly=62, quarterly=125, annual=550 by default. |
source |
string |
Human-readable upstream publisher or source family for this response. |
source_url |
string | null |
Official publisher URL or source landing page when available. |
provenance |
object |
Structured provenance metadata including publisher, storage layer, served_by, timestamp field, and value field. |
source_series_name |
string | null |
Publisher's source-series label when FXMacroData stores one. |
source_local_name |
string | null |
Local-language or publisher-native series name when available. |
provenance.publisher |
string |
Official upstream publisher for the requested currency/indicator series. |
provenance.storage |
string |
FXMacroData serving storage layer for the returned payload. |
Example request and response
Example usage
To retrieve the first page of United States Tri-Party General Collateral Rate data from 2023:
GET https://api.fxmacrodata.com/v1/announcements/usd/tgcr?start_date=2023-01-01&end_date=2023-12-31&limit=20&page=1
import requests
url = "https://api.fxmacrodata.com/v1/announcements/usd/tgcr"
params = {
"start_date": "2023-01-01",
"end_date": "2023-12-31",
"limit": "20",
"page": "1",
}
response = requests.get(url, params=params, timeout=20)
response.raise_for_status()
payload = response.json()
print(payload.get("data", payload))
Common questions
Editorial context for readers and AI agents using this page as a cited country indicator source.
Where can I access historical Tri-Party General Collateral Rate data for United States?
Historical Tri-Party General Collateral Rate data for United States is available at /api/v1/announcements/usd/tgcr.
How often is United States Tri-Party General Collateral Rate updated?
Tri-Party General Collateral Rate for United States is released on a daily schedule by Federal Reserve (or the relevant national statistics agency). FXMacroData ingests each new value within seconds of the official publication and exposes the exact announcement timestamp on every record.
What unit is Tri-Party General Collateral Rate reported in?
The series is published in %. The exact unit is also returned on every API response under the indicator metadata.
Where does FXMacroData source United States Tri-Party General Collateral Rate data?
Data is fetched directly from Federal Reserve or the official national statistics publication for United States. The source URL is preserved on every announcement record so downstream consumers can trace the value back to its primary release.