I suspect I'm just missing something in the manual explanation, but I don't understand the results I'm seeing when I enable Guyton-Klinger.
With Specified Expenses Only turned on, in Year of Retirement, my Essential Spending Rate is 4.96%, Non-essential is 3.14%, for an Overall rate of 8.09%.
Subsequent years, Overall spending is 7.63%, 6.44%, etc.
When I turn on Guyton-Klinger and set an initial rate of 8.0%, my YOR spending is E=5.05%, NE=0.0% for an overall rate of 5.05%. Subsequent years are 4.46%/1.22%, 3.18%/0.68%, etc.
My non-essential spending went to basically zero in YOR, and is a small fraction of what I had entered for Phased non-essential spending.
Is the G-K Spending Rate entered supposed to be Overall Spending?
Why doesn't my YOR Overall spending come close to 8%? Do I not understand what this should be doing? I thought it would adjust my non-essential spending to bring my overall spending to the initial spending rate, then adjust going forward.
Thanks for any help.
Todd
@tcbarney Hi, thank your for your question. I see the data in question in your scenario 2. I will email you with supporting data from your plan.
First, to address your question about the whether the Spending Rate applies to Overall Spending or not: GK spending is based on overall spending, unlike some of the other spending strategies for which, as I recall, the specified spending rate is applied to 'non-essential' expenses (a discussion for another time).
- the GK spending is calculated per the GK rules.
- the GK inflation, prosperity and cap preservation rules are applied, as applicable
- GK spending is $X which is 8% of PY savings, as expected (I confirmed this in your scenario 2, 2026)
- Pralana then subtracts your prior year essential spending (see note 2) to get a preliminary variable (non-essential) amount.
- It then applies a +/-20% limit (see note 1) on the year-over-year change in the GK spending.
Second, I see a bug in calculating spending in the retirement year: In that year, the prior year's GK spending is $0...since it is pre-retirement. Regrettably 120% of $0 is still $0. I will need to determine some appropriate way to address this, probably by using the prior year's total spending and have created a bug report for tracking.
Third, for the reasons noted below, basing this year's variable spending in part on prior year essential spending and applying year-over-year limits on the change skews the results so that the actual spending rate is not the expected 8%.
I do not recall that the published research on these spending strategies addresses a distinction between the concept of 'essential' and 'non-essential' spending. This introduces complications:
Note 1: Example: you specify 6% spending in one of these strategies, but in one or more years your essential spending is 8% for some reason...or there are big swings in essential spending from year-to-year. To handle this and protect the user from too much volatility in variable (non-essential) spending, we have applied an overall +/- 20% limit on year-over-year changes to the variable spending amount after applying any strategy limits.
Note 2: The ideal thing to do would be to use 'this' year's essential spending, not the prior year's. The problem is that this year's essential spending includes this year's taxes, which cannot be calculated until after all income and spending (which may include ACA subsidies, also dependent on taxes). Its a catch-22. So Pralana currently uses prior year essential spending.
I am sure our implementation of the spending strategies will evolve over time and look forward to discussions with interested subscribers about how to accommodate essential expenses. I am developing an iterative tax calculation (for same-year taxation of taxable unscheduled withdrawal) and that may present an opportunity to use current year essential spending in the GK algorithm.
Charlie, Pralana
Thanks, Charlie. It may take me a bit to wrap my head around your response.
In the meantime: Does the Retirement Year bug exist for other spending strategies as well? When I run Target % Adjustment at 8%, for example, my non-essential spending goes to 0% in retirement year, is 1.22% the following year, and then sits at 0% until death (with the exception of one year when I sell my house.) My overall spending hovers around 2.5%. I don't understand that result at all.
Fixed % Spending, I can sort of see what it's doing. In retirement year, there is 0% non-essential spending and overall spending of 5.05%, but then subsequent years it climbs back up towards the 8% target (though it does something funky in the 2 years after we sell our home.)
I'm not in a rush. What I actually want to do is model a Risk-Based Guardrail Strategy, but I wanted to understand where some of the other strategies would take me, first, in order to compare. I thought I knew, generally, what they would tell me and these results are throwing me for a loop.
Thanks,
Todd