Historical stress test

Each plan year takes three figures from its historical year, 1928 through 2024:

• Stocks: the U.S. large-company total return, including dividends. It is split into dividends and price growth using that year’s recorded dividend yield, so brokerage dividend income follows history too.
• Bonds: the U.S. Treasury bond return.
• Inflation: the December-to-December change in consumer prices.

Historical inflation replaces your inflation assumption everywhere it applies: rising expenses, blank-rate raises and Social Security increases, and federal brackets and limits after 2026. Your bond allocation, property appreciation and all entered amounts stay the same.

The mapping is one to one and in order. If your plan outlasts the data, the sequence continues from 1928. From that point the replay is a stitched scenario rather than one continuous period, and the result says so.

A partial first year compounds its historical return over only the months modeled, like any first year. See Your plan’s start date and first year.

The replay line is drawn against the steady-return forecast. Alongside it you get the lowest year-end net worth, the lowest year-end cash and investments, and the net worth difference from the forecast in its final year. The outcome reads Survived when spending is funded every year and net worth stays positive, Survived, but when spending is funded but net worth turns negative, and Failed when spending goes unpaid. All amounts are nominal dollars.

A replay that starts with a crash isn’t always the worst one. Here 1928 recovers while 1966 fails. From 1966 through 1981 prices nearly tripled, raising spending, while stocks and bonds both lost value after inflation. From 1928 through 1942 prices ended slightly lower.

A hypothetical preset changes one part of your plan from a start year you choose, then runs the steady projection. Every input and year the preset doesn’t touch keeps your own entries and assumptions. The event begins in January of the start year, or in the plan’s first month when that is the start year.

The market downturn preset spreads a stock decline over one to ten years from the start year, then lets stocks climb back. You set:

• Years of decline (1–10) and Total stock decline (%) (1–90): the cumulative peak-to-trough loss in stocks’ total return, dividends included.
• Recovery years (0–10): how long stocks take to return to their level before the downturn. Zero returns to your own assumptions the year after the trough.
• Bond return (%/yr) (−50 to 50) and Inflation (%/yr) (−10 to 30): annual rates for the decline years only.

Every decline year loses the same share, so the yearly losses compound to exactly the total you enter. Each recovery year earns the same return, chosen so stocks end the recovery back at their pre-downturn level. That is the level stocks started from, not where your assumed growth would have taken them.

Your bond return, inflation and dividend split assumptions stay in place through the recovery. In the decline years a bond or inflation rate you enter replaces your assumption: inflation then drives expenses, raises, Social Security increases and federal indexing for those years, as in a historical replay. Years past the end of your plan are dropped. When the downturn starts in a partial first year, that year’s whole share of the decline falls in the months modeled, as in the market crash preset. After a run, a chart traces the stock level through the decline and recovery.

• Changes within a year. Each year gets one annual return, so a low point inside a year can be lower than the year-end balances shown.
• Historical tax law, interest rates on your loans, or property values. Only the three market series change.
• Other markets, or histories before 1928.