Worst result of any calendar window from 1928 to 2025, before inflation, with each mix rebalanced every January and no fees or taxes (Exent calculation on NYU Stern’s annual returns; in brackets the window and how many windows lost money):
Starting in 1950 instead removes the Depression and softens the stock rows: 100% stocks lost 36.6% in 2008, 37.4% over 2000–02 and 11.1% over 2000–04, and 60/40 lost 18.0% in 2022, 20.8% over 1973–74 and 11.0% over 1972–74, with no losing 5-year window. Three warnings. Calendar years hide deeper drops inside them: the S&P 500 ETF fell 56.5% from peak to trough in 2007–09 (Exent market data), far more than 2008’s −36.6%. “Bonds” here means 10-year Treasuries, which lost 21.5% over 2021–22, so a bond fund with a long maturity is not a safe parking place for money due in two years. And none of this includes inflation, which matters less over 2 years than over 20.
The pattern for 2 to 5 years shows up in fund data too. In calendar 2022, a 1–3 year Treasury fund (SHY) lost 3.9%, a 7–10 year Treasury fund (IEF) lost 15.2% and a T-bill fund (BIL) returned +1.4% (Exent market data). Maturity, not the word “bond”, decides how much a rate jump can cost.