How much stress testing the retirement plan is enough?

[morning_miles]

How much stress testing the retirement plan is enough?

I’ve been using ******** to run retirement scenarios. I’m 49 and my wife is 47. I get 100% success with spending at 20k above my base budget. This is for retiring at end of year with a 40 year retirement. Extending out to 45 or 50 years doesn’t have a huge effect.

I feel confident with my base budget numbers. I’ve always had a household budget so the retirement budget is the actual current spend adjusted by how I envision retirement life (increased travel, increased entertainment expenses, etc.). I pay cash for vehicles and use a sinking fund so vehicles are covered in the budget. There is separate cash set aside for potential major home repairs.

I’ve tried to stress test the plan with what I see as the most likely threats.

• Elimination of ACA (my base budget includes health insurance premiums while keeping income below ACA income limits)
• High medical expenses before Medicare (both currently in good health)
• Reduction of social security benefits (in ******** I’ve used soc sec income per soc sec website at 70 considering no future earnings)
• Significant reduction in investments values (I read too much investment news and any AI bubble talk starts to make me nervous)

I believe I could absorb any one of the above threats. Two of the above could be absorbed, but it would start to eat into the travel budget. I have never traveled that much so I’m not sure if I would even want to in retirement, but it seems like that what people do so I’ve budgeted for it. Also, I have a bad back so I’m not even sure if I can. A combination of three of the above threats starts to be questionable. So my question is when does the stress testing start to become kind of ridiculous. (ie layering threat upon threat upon threat) How did you handle this in your decision of when to retire?

 

[RetiredHappy]

I think the key is flexibility. If your portfolio shrinks, can you cut expenses, i.e. not spend that $20K that you have above your base budget? ACA won't be eliminated as it will be political suicide. In our case, I simply don't qualify for ACA subsidies and I buy my plan off-exchange, and our annual medical expenses is about $30K to $35K between the two of us. We also have a concierge PCP which adds to our premiums and out of pocket costs. I haven't budgeted for reduction in SS benefits, but I know we will be fine even if they are cut by 30%.

You can also pad your expenses more to see if you still get 100% success rate.

 

[MichaelB]

So my question is when does the stress testing start to become kind of ridiculous. (ie layering threat upon threat upon threat) How did you handle this in your decision of when to retire?

Hi morning_miles. You’re probably already there, or not too far. This is just the financial part of retirement. There are other aspects, such as health and marriage, that also matter and you can’t model. We sometimes overemphasize the things we can measure and don’t think about the other stuff.

Once you've identified risks, what matters is not so much worse it can get but instead how will you react. Hard to model, but really important.

BTW, ER Fourm SW replaces some retirement calculator names with the *** because of past issues. Nothing to worry about.

 

[Al18]

Inflation is one of the biggest factors that can affect your retirement. In the past 7 years, inflation has increased prices by 29% (using the SS cola increases).

Also, I would be more conservative in your SS benefits. Instead of claiming at 70, assume you'll need to claim at 65 due to some event AND your benefits will be reduced by 25%.

 

[SecondCor521]

How much stress testing the retirement plan is enough?

I’ve been using ******** to run retirement scenarios. I’m 49 and my wife is 47. I get 100% success with spending at 20k above my base budget. This is for retiring at end of year with a 40 year retirement. Extending out to 45 or 50 years doesn’t have a huge effect.

One risk I feared was making a math mistake. You might try putting your situation into firecalc.com. I'd also suggest a baseline check - take your base budget and multiply by 25, and that should be around your FIRE stash amount (include an NPV for Social Security if SS is a big part of your retirement plan).

I feel confident with my base budget numbers. I’ve always had a household budget so the retirement budget is the actual current spend adjusted by how I envision retirement life (increased travel, increased entertainment expenses, etc.). I pay cash for vehicles and use a sinking fund so vehicles are covered in the budget. There is separate cash set aside for potential major home repairs.

I’ve tried to stress test the plan with what I see as the most likely threats.

• Elimination of ACA (my base budget includes health insurance premiums while keeping income below ACA income limits)
• High medical expenses before Medicare (both currently in good health)
• Reduction of social security benefits (in ******** I’ve used soc sec income per soc sec website at 70 considering no future earnings)
• Significant reduction in investments values (I read too much investment news and any AI bubble talk starts to make me nervous)

You might add high end of life medical costs to the list, although this doesn't really raise risk much, because the chances of you living that long and having high end of life medical costs and hit a bad patch historically are pretty much zero.

I believe I could absorb any one of the above threats. Two of the above could be absorbed, but it would start to eat into the travel budget. I have never traveled that much so I’m not sure if I would even want to in retirement, but it seems like that what people do so I’ve budgeted for it. Also, I have a bad back so I’m not even sure if I can. A combination of three of the above threats starts to be questionable. So my question is when does the stress testing start to become kind of ridiculous. (ie layering threat upon threat upon threat) How did you handle this in your decision of when to retire?

Two comments:

First, you'll never get the FIRE risk to zero. But there are risks to staying at work, too: you're spending time and energy on the job that could be spent elsewhere. Your job might be physically or mentally stressful. So a natural transition point would be when the risk of FIRE becomes less than the risk of working.

Second, we usually only model downside risk and worst case outcomes. But it is far more likely statistically that you'll end up above worst case, especially where you're modeling multiple downside risks combined to create a perfect storm of downside.

Anecdata, but I FIREd in 2016. I assumed a 4% withdrawal, 10.7% investment returns (which around here is regularly considered too optimistic), zero inheritance, zero extra income, 3% inflation, and ACA status quo. What in fact happened was 2% withdrawal, 15.5% investment returns, small inheritance in 2016 and large inheritance this year, $XK in side gig income, about 0.5% personal inflation, and the ACA has been improved in a few ways. Perhaps I have been luckier than average, but I made six planning assumptions and all of them turned out better than expected.

 

[WhenIsItTime]

Be helpful to know what $20k represents as a % miss. I preferred to use WR as our only income is from savings/investments. Our budget was near ~3% when I retired 7 years ago and <50 yrs old. Now it is at 2.2%. This includes spending as much (healthcare for family of 4 has doubled) or more than when working.

 

[Dtail]

It sounds like you have things covered. Try Firecalc or if you have access to the Fidelity calculator.

 

[morning_miles]

Be helpful to know what $20k represents as a % miss. I preferred to use WR as our only income is from savings/investments. Our budget was near ~3% when I retired 7 years ago and <50 yrs old. Now it is at 2.2%. This includes spending as much (healthcare for family of 4 has doubled) or more than when working.

Base budget = 85k.

 

[engineernerd]

How much stress testing the retirement plan is enough?
...

As much as you need to feel ready, is my answer. Not much fun to spend all your suddenly ample free time in a state of anxiety (applies to both spouses in a couple, but fearful spouse should try hard to self-educate so anxiety is based on real thought and can be discussed).

If you've been using a retirement calculator with a different philosophy than FIRECalc, I'd recommend running your scenarios in FIRECalc. And read the text in FIRECalc until you have a thorough understanding of what it's doing.

What I did was to run FIRECalc with my worst-case assumptions:

(1) I will live to 100
(2) my Social Security benefit will drop to 77% of its promised value in 2033 (per the 2025 Trustees' Report)
(3) my income taxes will be 20% of my gross income (income taxes have been higher in the past than now)

to see whether my basic living expenses would be covered.

Don't make assumptions about investment performance. FIRECalc includes actual historical portfolio performance data going back to 1871, along with concurrent historical inflation data, so it already has that base covered.

 

[TickTock]

Being able to absorb 2 of the 4 conditions by eating into your discretionary spending seems reasonable to me.

 

[morning_miles]

As much as you need to feel ready, is my answer. Not much fun to spend all your suddenly ample free time in a state of anxiety (applies to both spouses in a couple, but fearful spouse should try hard to self-educate so anxiety is based on real thought and can be discussed).

If you've been using a retirement calculator with a different philosophy than FIRECalc, I'd recommend running your scenarios in FIRECalc. And read the text in FIRECalc until you have a thorough understanding of what it's doing.

What I did was to run FIRECalc with my worst-case assumptions:

(1) I will live to 100
(2) my Social Security benefit will drop to 77% of its promised value in 2033 (per the 2025 Trustees' Report)
(3) my income taxes will be 20% of my gross income (income taxes have been higher in the past than now)

to see whether my basic living expenses would be covered.

Don't make assumptions about investment performance. FIRECalc includes actual historical portfolio performance data going back to 1871, along with concurrent historical inflation data, so it already has that base covered.

I believe the retirement calculator I'm using has generally the same philosophy as FIRECalc, but I ran my numbers thru FIRECalc just to confirm. I did get roughly the same result. 100% at 40 years, drops to 99 point something at 50 years.

 

[Al18]

If you are not comfortable with the numbers, rerun them every 6 months and decide.

 

[MichaelB]

I believe the retirement calculator I'm using has generally the same philosophy as FIRECalc, but I ran my numbers thru FIRECalc just to confirm. I did get roughly the same result. 100% at 40 years, drops to 99 point something at 50 years.

I think 50 years is just too long a period to be modeled by a calculator. Too much change takes place over that amount of time.

 

[engineernerd]

I believe the retirement calculator I'm using has generally the same philosophy as FIRECalc, but I ran my numbers thru FIRECalc just to confirm. I did get roughly the same result. 100% at 40 years, drops to 99 point something at 50 years.

Sounds reassuring to me...

If you want greater confidence about your portfolio composition being right, try out the "Investigate changing my allocation" option in the Investigate tab.

If you want a deeper understanding of what conditions give your portfolio the worst beating, try running your numbers in FI Calc and hunting down the worst retirement year in the year-by-year results. Then request spreadsheet output for that year from FIRECalc (under "Display the results of the retirement plan" in the Investigate tab).

For most stock allocations I find the worst year is 1966. The portfolio is harmed mostly by high inflation but also includes scattered negative-returns years. Stagflation, in a nutshell.