Macro data
Macro data, exactly as it was published
18,900+ US economic series back to 1854, plus Taiwan and South Korea chip exports. Every release and revision is kept, so your backtest sees only what was known on each date. Pay once for the history: one theme $29, every series $95, everything with every vintage $395.
Pricing
| Macro | What you get | API calls | One-time | |
|---|---|---|---|---|
| Free | Last 12 months of every series | 1,000 a day | $0 | Free key |
| One theme | Every series in one theme, full history, every vintage | No limit | $29 | Choose a theme |
| Macro Latest | All 18,900+ series, full history, latest values | No limit | $95 | Buy |
| Macro Complete | Every series and every vintage, first releases, the release calendar, money and liquidity, Treasury TIC capital flows, licensed series (Moody's, University of Michigan, Case-Shiller, ICE BofA, Cboe) and the Quantitative Model history | No limit | $395 | Buy |
Plus $19.95/month for new releases (one subscription for all your FinzData data; stop any time and keep the history you bought). With a look-back bundle: Macro Latest +$75, Macro Complete +$295. Institutional includes all macro.
Themes
Buy one theme for $29: every series in it, full history and every vintage.
Inflation & prices
1,139 series, from 1913. CPI, core CPI, PCE, PPI, import prices, commodities.
e.g. CPIAUCSL, PCEPILFE, PPIACO
Jobs & wages
804 series, from 1900. Payrolls, unemployment, average hourly earnings, claims, hours.
e.g. PAYEMS, UNRATE, CES0500000003
Rates & yield curve
325 series, from 1900. The Treasury curve from 1 month to 30 years, policy rates, spreads, financial conditions.
e.g. DGS3MO, DGS10, T10Y2Y
Growth & production
1,700 series, from 1854. GDP, industrial production, factory orders, trade, recession indicators.
e.g. GDPC1, INDPRO, USREC
Housing
300 series, from 1929. Starts, permits, sales, listings, inventory and construction spending.
e.g. HOUST, PERMIT, ACTLISCOUUS
Money, credit & liquidity
1,569 series, from 1921. M1, M2, bank credit, consumer credit, the Fed balance sheet.
e.g. M2SL, WALCL, TOTALSL
Consumer & income
7,618 series, from 1929. Spending, income and saving, nationally, by state and by country.
e.g. PCE, DSPIC96, PSAVERT
FX
106 series, from 1957. Dollar exchange rates and trade-weighted dollar indexes.
e.g. DEXJPUS, DEXUSEU, DTWEXBGS
Energy & electricity
1,024 series, from 1919. Electricity generation, retail prices and use.
e.g. Electricity series by state and sector
Asia tech exports
32 series, from 2000. Taiwan and South Korea chip, server and electronics exports, by product and destination.
e.g. KREXPIC, TWEXPICT, TWEXPUS
Why point in time matters
Economic numbers change after they are first published. Payrolls are revised twice, GDP three times and every year again. A backtest that uses today's revised figures trades on information nobody had at the time.
FinzData keeps every version with the date it became public. Ask for any series as of a date and you get exactly what was known then. For example, as of 1 May 2020 the latest US unemployment rate was March 2020 at 4.4%, the figure the market traded on, before April's 14.7% was released.
Get it in Python
import finzdata as yf
c = yf.Client()
# today's revised history
cpi = c.macro("CPIAUCSL")
# exactly what was published by 1 May 2020 (point in time)
u = c.macro("UNRATE", as_of="2020-05-01")
# every release and revision, or just the first print
all_v = c.macro("PAYEMS", vintage="all")
first = c.macro("GDPC1", vintage="first")
# Korea chip exports
chips = c.macro("KREXPIC")Also over the REST API (/v1/macro?series=CPIAUCSL&as_of=2020-05-01) and the MCP server for AI agents.
Sources
US series come from the Bureau of Labor Statistics, the Bureau of Economic Analysis, the Federal Reserve, the Treasury and the Census Bureau, with vintages from the St. Louis Fed's ALFRED archive and FinzData's own daily captures. Asia exports come from Taiwan's Ministry of Finance and Korea Customs Service. Third-party series carry their owner's credit on every download.
Frequently asked questions
What does point in time mean for macro data?
Economic data is revised after it is first published. A point-in-time series keeps every version with the date it became public, so a backtest only sees what was known on each day, with no look-ahead from later revisions.
What is the difference between Macro Latest and Macro Complete?
Macro Latest gives every series with its full history at today's revised values. Macro Complete adds every vintage (ask for any series as of any past date, or its first release), the release calendar, money and liquidity, Treasury TIC capital flows, licensed series and the Quantitative Model history.
What does one theme include?
Every series in that theme with full history and every vintage, for $29 once. Themes are inflation, jobs and wages, rates and the yield curve, growth, housing, money and liquidity, consumer and income, FX, energy, and Asia tech exports.
Is it a subscription?
The history is a one-time price. New releases keep arriving with the $19.95/month updates subscription, the same one that keeps all your FinzData data current. If you stop it, you keep the history you bought up to that day.
Is there a free tier?
Yes. A free API key gets the last 12 months of every series. No card needed.
I already have a look-back bundle. Do I need this?
Bundles include macro for the years you bought. Add Macro Latest (+$75) or Macro Complete (+$295) at checkout to unlock the full history of every series.