Investing - Theory, News & General • Why the 30-day T-bill *does not* earn its sobriquet as the risk-free rate

I've created a telltale chart from the FRED 3m Bill series. Here it is:



It tells basically the same story, but with Jan 2032 as the start date rather than Dec 1925.

I've added arrows to highlight the general trends: arrow pointing up, Tbills have a positive real return, arrow going down, negative real return.

Note that between Jan 1932 and Jul 1952 3m Bills lost about 46% in real terms. This is not what I'd call safe. Again, these periods of negative real return occur when the short-term rate is held low despite high inflation.

Notes on construction:

The 3m Bill actually is a 13w bill, which usually is 91 days, so I used 91 days as the term for the 3-month return calculations. So the growth for each series (bill and cpi) is calculated for each 3m interval; i.e., the compounding is every 3 months. The Treasury discount formula is used in calculating the bill growth factors, so there is a factor of 91/360 for each calculation.

Statistics: Posted by Kevin M — Sat Sep 26, 2026 2:32 pm