Retirement Lab
Research
Research notes
Original simulations on taxes, spending, markets, and the assumptions that change cross-border retirement outcomes.
Real markets crash harder than a bell curve predicts. We refitted how our simulator draws bad years, learned that how bumpy markets are matters far more than the shape of a crash, and changed the setting you control to match.
23 min readBengen's question still returns 4% on 150 years of data. Retiring at 40 on that same history needs about 24–29× spending; drawing years independently, with history's crash shape, needs about 33×; markets as rough as 1927–1956 push it to 46×; tax moves it as high as 50×; and the expected-return forecast moves it most of all.
16 min readAsk in plain language, get a real simulation: how to connect Retirement Lab to ChatGPT, six prompts to try, and where the app's limits are.
7 min readHolding purchasing power constant, the required portfolio ranged from 32.3× to 49.2× of local annual spending. No case reached the 95% target at 25×, and almost the entire spread between countries is modelled tax.
14 min readThe same household required $650K in the lowest threshold case and $2.73M in the highest. The ranking shows how national price levels and modeled tax treatment compound across a 55-year retirement.
12 min read