I would enter that as an annuity James. That way you can indicate the year it stops. (Pensions are assumed by the program to be payable for life). You can also use the annuity to indicate its growth rate. Since it's tied to the stock market you can just pick a growth rate you wish to assume that the stock market will sustain over the rest of the time period of the payout for the "annuity." This seems much more transparent and reliable to me than using Monte Carlo since you can try a few different assumed real return rates. Put 0% to merely match your inflation assumption (which of course probably around 2.5% or 3% nominal). Stocks might do better than that or much worse depending on the time period. I'd assume something conservative for my model if it were me. The MC would indicate a 6% inflation-adjusted return with a 15% standard deviation if you were looking at total stocks from the period of 1972 to present. The problem with using MC is that you don't know if that 45-year period is going to represent the next 45 years or not. For me, I'd just run one at 0% real, 1% and 2% and see what it looked like. Perhaps it matters a lot; perhaps it matters little on your discretionary spending.