Let's say I have a 1m investment portfolio today. And let's say my current year expenses is projected to be 40k. So I first withdraw 40K from the 1m, put the 40K to a money market account/checking account.
Now my portfolio is at 960k.
And next year comes along, assuming a 7% return, my portfolio is at 1.027m. I take 4.1k out. (assuming 3% inflation). And then repeat the process.
Let's say I am 45 and will live to 95. According to this calculation, I will actually have more than 2m left over for my kids when I die. Is this too good to be true? (see my fire-portfolio.ods attachment)
More or less, this is the idea.
However, it is important to note that, while the long term after-inflation returns of the stock market have been approximately 7%, you won't get exactly 7% every single year, and that matters. Specifically your equation is:
year 1: (x-y)*z
year 2: x2-y2)*u
^repeated long enough, you'd expect the averages (not simple arithmetic mean) of the z's and u's to approach 7%.
e.g.
year 1: (1 million - 40k) * z
year 2: (y1 result - 41k) * y
Sequence of returns risk, or SORR, is the risk that you get a few bad years right up front. Mathematically this matters because you have both addition and multiplication in there. For a simple extreme example, imagine you start with $10 and spend 1$ a year. The first year the market drops 80%. (10 - 1) * .2 = 1.8. The next year it goes up 400%, and that leaves you with (1.8 - 1) * 4 =
3.2. If you'd had the opposite order of years, you'd have (10 - 1) * 4 = 36, then (36 - 1) * .2 =
7. Even though the average return is the same over the time period in question.
The 4% guideline is attempting to take this into account to spit out a reasonable failure rate of a portfolio - if it didn't, it would just be the 7% rule =).
As you continue reading material, you'll find that most FIRE failures fail because of this (or because spending goes way up). Google Early Retirement Safe Withdrawal series to dive into the weeds. Keep in mind as you read though that people pay a lot more attention to the failure cases because the impact of running out is a lot more important than getting super rich. For instance, while your calculations showed ending up with 2m, nothing's to say you won't end up with 3m even without extremely good (historical) returns.
Different calculators use different approaches to attempt to deal this, but none are foolproof. And they all come with the important disclaimer that past performances are not indicative of future results -> 7% is not a mathematical law, just what we've seen in the past (in the US, in a period of global US domination, in a period of massive productivity gains for the planet, etc.).
That was the big picture question. The other question is about income and health care cost. So in order to get an affordable health plan from healthcare.gov, the portfolio needs to be set up to generate the right amount of dividends as income, is that right?
And the 40K yearly withdrawal needs to cover all the expenses including taxes for the dividend income, right? I would appreciate for some real numbers from people who are doing this.
Yes you need to include income taxes as an expense with the 4% calculation. The good news is that taxes will likely be very low on 40k withdrawals, especially since not all of that withdrawal may be subject to tax, e.g. only the gains in a taxable account, none of it in a Roth IRA. E.g. you might sell 40k from your taxable account, made up of 32k of contributions (not taxed further) and 8k in market gains (subject to tax). Or, 40k from your 401k which will all be subject to tax.
Dividends should be thought of as forced withdrawals for the most part when it comes to tax accounting, which is why many don't actually prefer them.