Please check my math/logic.

[Jerry1]

I entered in my quarterly results tonight and have a question. First a comment - I've had two good quarters. Hope it continues. But my question is:

Using dummy numbers. Say my starting retirement portfolio was $1M and since then to now it's $76K less. So, 7.6%. I've been retired 4 years and 1 quarter. If I divide 7.6% by 4.25 I get 1.8%.

It's not exactly my withdrawal rate since it includes everything (investment income/losses, distributions, income such as pension and SS and all expenses). Everything. Can I compare this to the 4% Trinity withdrawal rate analysis and assume that I'm doing pretty good? Or, is this just a completely different number. At any rate, it feels pretty good. Especially after the crazy last couple years and the selling I did at the low point of the pandemic . So, how am I doing 4.25 years in?

 

[ERD50]

Can I compare this to the 4% Trinity withdrawal rate analysis and assume that I'm doing pretty good?

No.

Or, is this just a completely different number.

Yes.

(more to follow...)

-ERD50

 

[ERD50]

Can I compare this to the 4% Trinity withdrawal rate analysis and assume that I'm doing pretty good?

No. The 4% withdrawal rate means 4% of your initial portfolio, and increased by inflation every year.

Or, is this just a completely different number.

Yes. Simple math, first year:

$1M portfolio, 4% is a $40,000 withdraw amount, leaving $960,000 if there were no portfolio gains or losses.

But your investment may have gone up XX%, or down XX%. So there's no comparison.

Look at a FIRECalc default run, the numbers in the first few years are all over the map, but the WR is consistent. You don't know for 30 years!

If your drop is as bad as one of the failures, that could be cause for concern, or maybe not - the following years might be better. Who knows?

-ERD50

 

[ERD50]

OK, a real analysis:

https://tinyurl.com/2ossf6vu << short link to portfolio-visualizer

From Jan 2019 to EOM Feb 2023

$1M portfolio, 70/30 VTI/BND

$40,000 spend, inflation adjusted.

start; end: $1,000,000 $1,274,037

So no, by that measure, if your spending really is ~ 4%, you should be up 27.4%, not down 7.6%. And that doesn't account for any pension/SS, which would effectively reduce your withdrawal amount, leaving even more $ in the end number.

Sorry!

-ERD50

 

[Jerry1]

OK, a real analysis:

https://tinyurl.com/2ossf6vu << short link to portfolio-visualizer

Thanks. That helps. I took that link and did a couple things. First, I adjusted it for my actual asset allocation which is closer to 30/70 (equities/fixed). That pretty much broke even on a $40K withdrawal. Ending number was $1,016,788.

Then I adjusted the withdrawal amount until I got to my 7.6% reduction. That required me to enter $62,500. I assume it's fair to say that I'm about where I would expect if I started with a 6.25% withdrawal rate.

Thankfully, that's not bad. Reason being is that in 2023, both DW and I started SS. Therefore, actual withdrawal rate will decrease substantially. SS will cover about 55% of my total spending so all else being equal my hypothetical withdrawal rate going forward should be closer to 3.4%.

In actuality, my pension and our SS pretty much cover our spending so my withdrawal rate should be very low. Though I clearly need to do better with my portfolio return. (Okay, I'll do that. )

 

[ERD50]

Thanks. That helps. I took that link and did a couple things. First, I adjusted it for my actual asset allocation which is closer to 30/70 (equities/fixed). That pretty much broke even on a $40K withdrawal. Ending number was $1,016,788.

Then I adjusted the withdrawal amount until I got to my 7.6% reduction. That required me to enter $62,500. I assume it's fair to say that I'm about where I would expect if I started with a 6.25% withdrawal rate.

Thankfully, that's not bad. Reason being is that in 2023, both DW and I started SS. Therefore, actual withdrawal rate will decrease substantially. SS will cover about 55% of my total spending so all else being equal my hypothetical withdrawal rate going forward should be closer to 3.4%.

In actuality, my pension and our SS pretty much cover our spending so my withdrawal rate should be very low. Though I clearly need to do better with my portfolio return. (Okay, I'll do that. )

That sounds closer to reality. Also, recall that the Trinity Study, and the FIRECalc defaults are based on 75/25 or 70/30 AA something in that range.

Your 30/70 AA has historically had more failures (87.7% success) than a 75/25 (95.0% success). Generally, the growth from stocks, tempered by some fixed income, provides a higher portfolio success.

-ERD50

 

[Jerry1]

Thanks for your input.

Another thought overnight. When I take my actual annual reduction and divide it by my starting value, I get 52 years. That seems pretty good especially since we are now on SS. Also, my “number” is my investable assets so that doesn’t include our house.

FWIW, we’re revisiting our overall financial plan and looking to put something in place to do better and have less moving parts (less room for me to screw up). DW is finally going to move her 401K to Fidelity. She had a 3% fixed rate account with most of her money in it so there was no need to move it until now, when the interest rates increased. We’re also looking to say goodbye to our FA. He’s been a good guy but the value is just not there. Hard decision because we like him a lot. The portfolio visualizer will be helpful in looking at setting up our future asset allocation.

 

[ERD50]

Thanks for your input.

Another thought overnight. When I take my actual annual reduction and divide it by my starting value, I get 52 years. That seems pretty good especially since we are now on SS. .....

Nah, there's no relevance to 52 years. You are just not thinking of this in a useful way.

It is unlikely that the next X decades will look anything like what has happened in the past 4 years. It just doesn't work that way. We've had four years in a market that is up despite withdraws. We will likely see some deep dips again in the future, maybe long periods of flat markets and/or crazy 80's style inflation.

What is relevant is, after SS and pensions, if you only need to pull a small % (well below 4%) from your current invest-able assets, you should be fine, if history is even close. And it appears you are. But the other measures are meaningless. You could put your current #s into FIRECalc or other tool.

Also consider other 'lumpy' spending, new roof, HVAC, car, etc.

Also, my “number” is my investable assets so that doesn’t include our house. .....

I think most people take this approach. You can't easily 'spend' your home. And if you sell it, the proceeds will go towards living expenses elsewhere. Unless one has a specific, actionable downsizing plan, just leave it out of the equation, IMO.

-ERD50

 

[RunningBum]

The past is past. Now that you are on SS, and have had 4.5 years to hopefully better estimate your expenses in retirement, does your plan look solid now? What does Firecalc say?

 

[Jerry1]

The past is past. Now that you are on SS, and have had 4.5 years to hopefully better estimate your expenses in retirement, does your plan look solid now? What does Firecalc say?

Haven’t run it against FireCalc. I’ll do that, but it will show success because my withdrawal rate going forward will be very low. I’ll do it anyway just because it will help me get my thoughts together. I’ll use it to model out some worse case spending scenarios like needed a roof, water heater and a new car in a short period of time.

 

[RunningBum]

Haven’t run it against FireCalc. I’ll do that, but it will show success because my withdrawal rate going forward will be very low. I’ll do it anyway just because it will help me get my thoughts together. I’ll use it to model out some worse case spending scenarios like needed a roof, water heater and a new car in a short period of time.

I'm just over halfway through a 20 year gap between ER and collecting SS, if I wait until 70 as planned. 15 year gap to a small pension too. It would have cheated me on spending had I not somehow included those things for those 15/20 years.

So I calculated the NPV of my SS (with a reduction in case benefit cuts are made) and pension and added them to my investment net worth, and set a spending target of X% of that value. You could alternately use a side fund to make up for that gap during that time. One way or another I think SS and pension should be factored in before you start taking them. This also keeps you from deciding to take SS as early as possible so you can start spending that money. There are good reasons to take SS early, but that isn't one of them.

So you could go back and factor the pension and SS in at the time you retired, and see what your WR rate really was since then. I'm guessing you'll find it was much safer than you have been thinking it was.

Your mention of the lumpy expenses is a good point too. If you didn't have any large irregular expenses in that short time, you could be lulled into a false sense of security that all is well, when the dam is about to burst on some big expenses. Or if you did have a lot of those, you could be worrying about a broken plan, when really things will even out over time as you may not have another big expense for years.

VPW helps with the "How am I doing" question because it starts adjusting your spending target up or down based on your start of year net worth, and it smooths out the increases and decreases over time. You still have to watch for that false sense of security if you've been underspending because you didn't have any of those large irregular expenses. And even though you "can't" run out of money because you're always targeting a less-than-100% of net worth spending, you could be down to $100 in your 90s and that probably won't work.

Keeping those limitations in mind, I use VPW because it takes the guesswork out of what to do after a down year like 2022. The Trinity study would suggest you can keep spending 4%+inflation, which probably works in most cases, but if the following year is bad and so is the next, you'll probably have to decide when to start cutting, and it's going to be more drastic.

 

[Jerry1]

Thanks. I'll look into that. As for SS, both me and DW have already started. Long story, but it was right for us. Short story is I have health issues and there was no point in waiting.

HaHa - Yes, the first four years have been very lumpy. On the plus, I won $50K in the lotto about two months into retirement. On the out flows, we installed an inground pool and bought two new cars. I certainly don't think the first four years are a model for the future, but it does seem like there is always something.

I put some information into FireCalc and it shows 100%. It stays that way until I add just over $20K extra in spending (per year). So, there seems to be a good amount of cushion. However, just moving the inflation number up to 5% causes the whole thing to crash. I have to drop spending about $30K to get back to 100%. That includes the $20K (the high point) so that's about $10K less than my current budget. That's a bit scarry given this past year of inflation.

Anyway, it was a good exercise. I'll look at the other things you mentioned. Thanks.

 

[SnowballCamper]

I think the math/logic of it is that math/logic can't predict the future. It is a useful tool to model possible outcomes such as with monte-carlo simulation... but this requires some assumptions that most of us aren't good at understanding. And then the question, "But has that ever happened?" comes up, and the answer is no. As a result we take the mental shortcut of looking at past performance...in the form of FireCalc around here.

Since there really isn't an objective answer to the question, "Do I have enough to retire (or stay retired)?" I think it's useful to consider how many years of expenses one has on hand in the retirement funds. This certainly doesn't answer the question, but it shifts the thought to the two essential parts of the problem.

Will my retirement funds grow at least at the rate of inflation to fund me this many years?

Will my retirement funds grow above the rate of inflation to fund my expenses if I live longer than this many years?

 

[ERD50]

... However, just moving the inflation number up to 5% causes the whole thing to crash. ....

What do you mean "moving the inflation number up to 5%"? You must not be using the default historic analysis that FIRECalc does, you must be choosing the MonteCarlo option?

For me personally (and I'm not alone), I don't use the MonteCarlo at all. It is all dependent on what you choose. I think history is a better guide. I like to throw in some margin (keep spending below the 100% level) to assure I could handle a future even worse than the worst of the past.

Remember, in the 80's inflation was double digits for years.

Remember, the FIRECalc success rate isn't about averages, 100% means you survive the worst of the worst in history, A future failure means the future is worse than the worst of the worst - which can happen, records were made to be broken. But that's a pretty steep hill to climb. And at any rate, you'll be better off than those who were not so conservative (assuming one can afford it, the other choice is work until you die!).

-ERD50

 

[Jerry1]

What do you mean "moving the inflation number up to 5%"?

In the Spending Model tab, the default is CPI. However, one of the selections is to choose your inflation rate. It shows as 3% but it can be changed. I changed it to 5%. I did not use a Monte Carlo simulator. I just used the constant spending model in FireCalc.

 

[ERD50]

In the Spending Model tab, the default is CPI. However, one of the selections is to choose your inflation rate. It shows as 3% but it can be changed. I changed it to 5%. I did not use a Monte Carlo simulator. I just used the constant spending model in FireCalc.

OK, I never played with that. But that still seems to me to be defeating the purpose of using historical data - consumer inflation is not static, just as market returns are not static.

-ERD50

 

[Jerry1]

OK, I never played with that. But that still seems to me to be defeating the purpose of using historical data - consumer inflation is not static, just as market returns are not static.

-ERD50

Do you know what it uses for inflation when you select the default - CPI?

 

[pb4uski]

I think it uses the actual CPI for that historic year, ditto with returns, etc. So each little line that you see is the starting portfolio as of the beginning of a year... say 1960, increased for 1960 returns and then withdrawals increased for 1960 inflation, etc... then 1961 and so on and so forth for the number of years that you specify.

 

[ERD50]

Do you know what it uses for inflation when you select the default - CPI?

To be certain, I opened it in a 'private window', and it had CPI checked (as I assumed).

And to be clear, it's using the historical values for each run. For example, for the 30 year cycle starting 1966, CPI for each year 1966 through 1996, CPI for each year is applied sequentially to the withdraw amount. No averaging or anything. The run for 1936 is going to use CPI for 1936 through 1966, etc. For the total number of runs tested.

edit/add - I cross posted with pb4uski - what he said

-ERD50