US Treasury Auction Internals

Every US Treasury auction since 2018, with its demand internals kept intact - bid-to-cover, high (stop-out) yield, coupon, the indirect / direct / primary-dealer take-down split, the tail/allocation percentage, and the size accepted - organized as a long panel of tenor × metric × auction date. No daily-mean collapse: this is the auction-by-auction read on who is actually buying government debt and how aggressively, tidied from the Treasury's paginated auctions_query into one clean Parquet.

What's included

• Per-auction demand internals (NO daily mean), every Treasury auction 2018-present
• zone = tenor + type (e.g. '10-Year Note', '4-Week Bill', '30-Year Bond')
• metric = bid_to_cover / high_yield / coupon_rate / indirect_pct / direct_pct / primary_dealer_pct / tail_alloc_pct / total_accepted_bn
• Bidder buckets converted to percent of total accepted; high yield/coupon in percent
• Long format - pivot to one row per auction, or track a single tenor's take-down over time
• Source: U.S. Department of the Treasury (public domain)

Schema

Sample & preview

Every purchase ships as Apache Parquet with a data dictionary and the full QC report. A free sample (first rows + schema) is downloadable here - confirm fit before you buy. Source: U.S. Treasury (redistributable open data; attribution passes through - see our Licence).

import pandas as pd
df = pd.read_parquet("us-treasury-auctions_sample.parquet")
df.info()   # typed columns, gap-aware, ready to join