Last Updated: February 7, 2023
Well I’ll admit this is a bit of a disappointing post for me, because I had really hoped to complete a major overhaul of the FIRE Withdrawal Strategy Algorithms Page I’ve been working on all week, as I mentioned at the bottom of my last post. But that’s how it goes sometimes – I’d rather have a decent product than rush to “ship” a super rough / sloppy product. Especially for THIS product – I expect this page will serve as an important reference for much additional analysis I’m planning.
The graphical diagrams on that algorithms page also need a big overhaul, to reflect the current version of the methods.
Why the Overhaul
I haven’t updated the algorithms page since November 8 of last year, and I’ve made a tremendous number of upgrades to the Traditional Method and especially the Tax and Penalty Minimization (TPM) method since then.
Some of the big changes include:
- including RMDs
- a revamp of dividends calculation
- enabling both methods to go down to $0
- no more iteration with Taxes
- many bug fixes
New Organization – Algorithms Vs Tool
As I began the overhaul, I realized that the current FIRE Withdrawal Strategy Algorithms Page is actually more a description of how the portfolio projection tool I’ve built implements the algorithms, rather than the algorithms themselves.
I then decided that what I really want on the Algorithms page is just the Traditional Method and TPM method that someone could follow each year of their retirement when making their portfolio withdrawals.
Thus I’d like to describe how the portfolio projection tool works on a new, separate page. So that’s what I’ll work on after updating the algorithms page.
Updating the GitHub Code
Something else that I hadn’t updated in far too long: the python code of the portfolio projection tool on GitHub. So that’s now fully updated.
Other Future Work
As usual I have near-infinite other things to work on next. Some of the top items are sorted by scope below.
Smaller Scope
With the passage of Secure 2.0, Required Minimum Distributions (RMDs) now start at age 73 or 75, depending your age in 2033. So I need to create a user input for the portfolio projection tool that allows the user to specify what age RMDs should start, including for a spouse if doing a projection for a married couple (which could be different for each person, depending on what age they are in 2033).
It’s also 2023 instead of 2022 – which means I need to update the various tax bracket values I currently employ in the portfolio projection tool. I’m not going to re-run all my analysis with these new numbers though, for a couple reasons: a) that would be a ton of work, for likely little benefit, and b) I expect the broad trends and conclusions of the analysis to hold for future years as bracket values tend not to change dramatically (typically just an inflation adjustment).
Medium Scope
Something I think would be really neat, and would be a nice way to hold my wife and I accountable in our finances and lifestyle, is to publish our expenses each month. Or if that’s too much, perhaps at least once a year publish our annual expenses. I’m going to start with a 2022 Expenses post.
I already do extensive tracking of our expenses on a monthly and yearly basis, so I hope (probably naively) that it won’t be too much more work to throw it up on the site as a blog post after I update our numbers in our spreadsheets each time.
I’d also love to show how we track our expenses, and perhaps also provide some templates others can use if they’d like.
I also got a nice idea from @minnesotafinances when we were chatting on Instagram: describing what we feel was worth spending money on. And I also like the opposite idea: describing purchases that we feel were a total waste of money. So hopefully I can do that as well.
Larger Scope
There are probably two main big projects I’d like to work on next, though I’m torn on which to pursue first.
ACA Subsidies
Assuming you’re FIRE and thus not working AND not eligible for Medicare, odds are you’ll be getting your health insurance via the Affordable Care Act (ACA) (“Obamacare”) health insurance exchanges.
And if you have Obamacare, the amount of subsidies you receive, and thus how much you pay for health insurance, is heavily dependent on your income – along with number and age of all people on the plan, your zip code, etc.
So that means that if your income rises, your subsidies go down, and your health insurance cost rises.
I would REALLY like to build and add an ACA subsidies / premiums model to the portfolio projection tool, so that I can see the impact of the subsidies on the FIRE portfolio performance and on the TPM and Traditional withdrawal methods.
In particular, I’d like to answer the following questions:
- What is the impact of increasing your LT cap gains beyond your targeted overall income? I know that will decrease your subsidies, but you may also be able to harvest for LT cap gains, so what effect is more powerful?
- Can the new withdrawal method I tried, which failed to improve performance in all scenarios I tried, finally prove valuable if I account for ACA subsidies? In particular, if the method can keep your income steady while delivering more cash for expenses, would the savings on health insurance premiums make the new method worth it?
Social Security – When to Start, and Impact of FIRE
I’d also REALLY like to return to social security income, which the Social Security Administration refers to as retirement insurance benefits (RIB). Because already writing two previous articles about it is clearly not enough!
All the analysis I’ve done so far assumes you take RIB at the standard full retirement age of 67, but I’d like to do some analysis for a wide range of scenarios to see whether you might want to take RIB earlier or later. And I’m particularly interested in the impact your start date/age has on spousal benefits.
A post about when to start RIB has been my number one request for a future topic so far. But, it’s going to require that I code up the method for how RIB is computed, including for spouses – not sure how easy that’ll be to track down and implement.
Once I have that “Compute RIB” method in place, something else I’ve wondered about for a long time is how retiring early via FIRE impacts RIB. E.g., if you retire at age 40 or 50 instead of 65 or 70, how much smaller can you expect your RIB to be? And how does that impact your time to FI during your saving phase?
It’s Late
Alrighty, it’s getting late, I better get this posted. As usual, my simple “Update” post has turned out much longer than I expected.
As always, let me know if you have ideas/opinions on future topics.