My basic profile is very simple: Among other things it assumes my nominal investment return on qualified and unqualified money matches 3% inflation (i.e. zero real return), assumes my wife and I both die at age 100, and inputs as special receipts the payouts of three immediate annuities, two of which are currently paying out, and the third of which is due to start paying out in June of this year. I don’t assume a house sale, insurance, leaving a legacy, or anything else to complicate matters. I plan to retire at the end of 2019.
I have the option of delaying the start of payout of the third annuity anywhere up to 5 years, if I make the decision within the next few weeks. I spoke with the annuity company (New York Life) and found out what the payout would be if I delayed the start for 2 years, or for 5 years (obviously higher in both cases), and found out in each case what portion would be taxable for the first 14 to 16 years, until the non-taxable component added up to my initial premium, at which point everything would be taxable going forward.
I combined all the taxable and non-taxable payouts on an Excel spreadsheet for each scenario (start third annuity payout in 2018, in 2020, and in 2023). I then made two copies of my basic profile, cleared the special receipts grid, and entered the proper numbers for the two delayed payout scenarios, being sure to enter both taxable and nontaxable receipts, and being sure to use dollars and not today’s dollars (since there is no inflation adjustment). So I now have three basic profiles, differing only by the start date of the third annuity.
When I ran all three reports, the suggested spending for the three scenarios differed by less than a tenth of a percent, i.e. almost identical. So if we both live to exactly 100, and our real investment returns are zero every year, it doesn’t matter which option we choose.
Then, in each of the three scenarios, I changed both our deaths to age 95. In the first scenario (third annuity start date 2018), the suggested annual spending DROPPED by about 1.5%, which seems awfully counterintuitive. In the second and third scenarios, the suggested annual spending increased by around 20%, which seems more reasonable.
Obviously I’ve done something wrong, but I’ve checked everything carefully, and can’t see a problem.