Early access is open. Free sample datasets are live now. Download a sample

Data product · Early access

Economic Data API: CPI, GDP, Jobs and Yields

The macro numbers that move every asset class, CPI, payrolls, unemployment, GDP, the fed funds rate and Treasury yields, as clean time series in the same format as your price data.

Coverage

What you get

Item Detail
Product code X-ECON
Coverage Public series from the BLS, BEA, Federal Reserve, Treasury and ECB
Granularity Daily, weekly, monthly, quarterly or annual, by series
History Full history as published by each agency. Release-date (point-in-time) history on Starter
Delivery REST and free bulk CSV or Parquet downloads
Update speed After each agency release

Most economic APIs repackage FRED and stop there. That gives you today's revised numbers, which is a trap in a backtest: the first payrolls print you'd have traded on is not the revised figure you see now. On Starter we add release_date, so you can test on what was known at the time.

Schema

Schema

Field Type Required Description
symbol string yes Series ID, e.g. CPIAUCSL, DGS10; yield curve uses UST-10Y style
name string yes e.g. CPI, all urban consumers
date string yes Period the value refers to
value number yes The number
unit string yes e.g. percent, index 1982-84=100
frequency string yes daily, weekly, monthly, quarterly, annual
country string yes ISO code, e.g. US
seasonal_adjustment string optional SA, NSA
tenor string optional Yield curve only, e.g. 10Y
release_date string optional When this value was published (point-in-time backtests)
source string yes e.g. U.S. Bureau of Labor Statistics

Example record (illustrative):

{
  "symbol": "CPIAUCSL",
  "name": "CPI, all urban consumers",
  "date": "2026-10-01",
  "value": 337.476,
  "unit": "index 1982-84=100",
  "frequency": "monthly",
  "country": "US",
  "seasonal_adjustment": "SA",
  "release_date": "2026-11-12",
  "source": "U.S. Bureau of Labor Statistics via FRED"
}

The free samples use the shorter layout symbol, series_name, date, value (indicators) and symbol, date, tenor, yield_pct (yield curve).

Sample data

Sample data

US Economic Indicators. Treasury yields, the yield spread, fed funds, CPI, unemployment, payrolls and GDP. First 10 rows:

symbol series_name date value
DGS10 10-year Treasury yield 2026-05-27 4.48
DGS10 10-year Treasury yield 2026-05-28 4.45
DGS10 10-year Treasury yield 2026-05-29 4.45
DGS10 10-year Treasury yield 2026-06-01 4.47
DGS10 10-year Treasury yield 2026-06-02 4.46
DGS10 10-year Treasury yield 2026-06-03 4.49
DGS10 10-year Treasury yield 2026-06-04 4.47
DGS10 10-year Treasury yield 2026-06-05 4.55
DGS10 10-year Treasury yield 2026-06-08 4.56
DGS10 10-year Treasury yield 2026-06-09 4.53

Source: FRED, Federal Reserve Bank of St. Louis

US Treasury Yield Curve. Daily par yields from 1 month to 30 years for the last 20 trading days. First 10 rows:

symbol date tenor yield_pct
UST-1M 2026-11-18 1M 4.1
UST-1.5M 2026-11-18 1.5M 4.24
UST-2M 2026-11-18 2M 4.27
UST-3M 2026-11-18 3M 4.34
UST-4M 2026-11-18 4M 4.35
UST-6M 2026-11-18 6M 4.43
UST-1Y 2026-11-18 1Y 4.62
UST-2Y 2026-11-18 2Y 4.8
UST-3Y 2026-11-18 3Y 4.84
UST-5Y 2026-11-18 5Y 4.89

Source: U.S. Treasury

Use cases

Use cases

Macro regime filters. Compute CPI year-over-year from CPIAUCSL and flag months where inflation is rising. Run your equity model only in the regime where it historically worked.

Point-in-time backtests. On Starter, filter rows to release_date on or before each backtest date. Your signal then sees the payrolls number traders actually had, not the revised one.

Rates context for any trade. Pull DGS10 daily and the fed funds rate next to your stock or FX bars. Joining on date takes one line, since every series uses the same date format as our prices.

Release reaction studies. Pair each CPI date with its release_date and measure how SPY or EURUSD moved in the hour after. Event times are in the economic calendar API.

Code

Code examples

Example: monthly CPI since 2020.

curl "https://api.marketdata.co/v1/economic/series?symbol=CPIAUCSL&start=2020-01-01" \
  -H "Authorization: Bearer $MARKETDATA_API_KEY"

Example: Python, the 10-year yield from the free sample with a 20-day change.

import pandas as pd

econ = pd.read_csv("us-economic-indicators.csv", parse_dates=["date"])
y10 = econ[econ["symbol"] == "DGS10"].set_index("date")["value"].sort_index()
chg = y10.diff(20)
print(pd.DataFrame({"dgs10": y10, "chg_20d": chg}).tail())

Pricing

Pricing

Plan Economic data
Free, $0 All public series, full history
Starter, $29/mo Adds release-date (point-in-time) history

Every public series is also a free bulk download. Among the providers we studied, economic data costs $0 to $1,250 a month.

Side by side

How we compare

13 of the 50 providers we studied sell economic data, mostly as a repackaging of FRED. We pull from the original agencies, give the series away on every plan, and use the same symbol and date layout as our market data. The one thing we charge for is release-date history, which none of the providers we studied advertise.

FAQ

Frequently asked questions

Are values seasonally adjusted?

It depends on the series, and seasonal_adjustment tells you which: SA or NSA. Where an agency publishes both, both are available under different symbol values.

Do you keep revised and original values?

The free series hold the latest published values. Point-in-time history with release_date, which keeps what was known on each date, comes with Starter.

What does date mean for monthly data?

It is the period the value describes, so CPI for October 2026 has date 2026-10-01. The day it was published is in release_date.

Is this data free to use commercially?

The source series are public, and we credit each agency. Some series on FRED carry third-party copyright, which is one reason we pull from the original agencies.

Is it real-time?

Series update after each agency release. For exact release times, use the economic calendar API.

What formats can I download?

JSON from the API, and CSV or Parquet bulk files. The free samples come as ZIP files with CSV and JSON.

Keep reading