What's a good way to arrive at your number?

[Yegey]

I am curious the method that one would arrive at the retirement number. My thought were

1. Estimate your current expense and probably predict any future creep.
2. Get an estimate of SS.
3. Subtract your SS and other retirement income (like a pension) from the expense.]
4. Estimate 25x for a 30 years retirement.
5. Run some calculation using different methods to check how realistic the plan is. I think typical recommendation is about 90%.
   • A direct method using a pessimistic, and average.
   • A historical; calc like firecalc or ********.
   • A Monte Carlo like in portfolio Visualizer or Fidelity Retirement Planner.

However, things never go as plan. I am curious about the following.

1. How do you avoid underestimating expense? My thought was that expenses must be recalculated every couple of years.
2. How do you handle that SS estimate will vary depending on when you file? I have known people that are forced to retire early and had to file early. My current mitigation is to be able to at least survive with no SS.
3. 25x expense is a good number for a 30 year retirement. What would be the expense for a longer or shorter retirement?
4. Do people even find the calculator useful? They seems to be very input sensitive. Calc result often vary a lot because they have different assumptions on inflation rate, tax treatment, etc. I suppose investigating their difference can dig up assumptions that you can see if you need to incorporate into your model.

Not everyone wants to be this detail. What's the simplest plan you can think of?

 

[Aerides]

Depending how old you are now, you might want to dial back the SS factor or not include it. We didn't, as we were retiring in our 40's. We'll consider it gravy when we get there.

You don't estimate expenses, you use actuals, going back a couple of years at least. And then you pad, and pad some more, and think about things like a new roof, car, healthcare, taxes, and pad again. And then 10 years later you find you want to travel more, and are glad you padded that 8th time.

25x for 30 years is the standard 4% rule, but it also includes zero withdrawals for the first year, and is requires some higher risk tolerance if you are retiring in a bumpy economy. For many it works just fine. We probably don't hear much here for those for whom it did not. (I would not have wanted to retire in 2006 with a 4% plan.)

If I wanted to go super simple and have no stress and retire at 40 today, I'd probably go with 40x-50x expenses, only if I were super sure about my expenses.

 

[Montecfo]

Yes need actuals. And then you want an estimate for "lumpy expenses" such a new car, new roof, new appliances, remodeling whatever.

When I was in planning mode I looked back 10 years in those categories in Quicken, then estimated an annual average.

We thought of SS as our longevity insurance (I've saved). But begin with some sort of logic and then you can improve and shape your estimates.

 

[imjustawarrior]

If one runs the calculator and sees that the outcome is sensitive to the input, then that means you are sensitive to the input. The calculator is doing its job. I used Firecalc and Fidelity retirement planner to determine when I was FI.

Estimated expenses for the rest of your days is one of the most important factors. I believe it is important to track expenses every year to rerun the calculators, especially if you want to use up all your money (spend, give away). If you are not retired, take a look at health insurance costs and don't forget to include taxes and one off expenses. Our spending has been in line with our estimates. Our lifestyle has not changed in retirement.

When you take SS is supposed to be actuarily neutral. In general, when one takes it has little impact.

I find the 25x expenses to be pretty useless compared to the calculators. Our spend from the portfolio is not constant. The 25x method does not handle that very well.

 

[SnowballCamper]

How do you want to manage the risk of running out of money?

If you're going to build a TIPS ladder then you'll need the full living expenses for however long you want your retirement to be funded. I think this is the simplest, but based on history it's the most expensive (because you don't get any stock market gains).

If you're using a diversified stock portfolio, then Firecalc is the most popular answer you'll get here. And yes, any model that projects over 30 years or more is going to be very sensitive to the inputs because of exponential growth.

 

[CO-guy]

We arrived at our FI number without any consideration for SS. If the system is still intact in a decade (me) or two (her), then we'll consider it a bonus.

 

[Dtail]

I think you have hit on the correct concepts overall.
I am one who puts a decent amount of faith in the retirement calculators with Firecalc and Fidelity my main ones.
Yes the inputs matter of course, but keep in mind that one can always course correct along the way. Another concept one can use with the calculators is once retired, one can run the calculators based on current portfolios with one's withdrawal rate.
Keep in mind that if one uses a 95% success rate, then there are 6 failures. None of those 6 failures' starting retirement year have happened in 60 years.

 

[RetiredHappy]

Your process is spot on, subtracting SS and pension first and then use 25X of the net expenses for retirement. I like Fidelity Retirement Planner as I really dislike the FIRECalc interface and lack of ability to put in alot of details of the expenses, timing (period, start and stop), etc. I simply do not use FIRECalc beyond trying it out initially.

I would think estimating growth in a portfolio is the hardest piece of the equation. What you invest in influences the future portfolio size. I think both calculators presume a moderately aggressive growth portfolio.

 

[Steelpony]

https://www.investor.gov/financial-tools-calculators/calculators/compound-interest-calculator

I use a compound calculator and use a 3% personal inflation rate which is partially covered by SS.

We have a set aside for later life issues if needed, VTI, on reinvestment. I track 3 LTC facilities yearly for current fees and adjust this set aside separately.

In summary I check everything near the end of every year with the most current facts.Then I might tweak any part of a portfolio if needed.

Otherwise I stay away from the daily news and the wall of woe and try to spend reasonably. The collateral benefit so far is excess income to needs due to compounding and DCA with no spend down after 16-17 years. This provides plenty of slop.

I really don’t have a number because there is no holy grail answer to an unknown.

 

[Yegey]

Note, when I said estimate of expense, it is actual but in reality expense may not be the same at retirement so it is at best an estimate. Hopefully it won't change much.

One factor I like to consider is flexibility in spending. A plan failure may just need a reduction in withdraw to fix. The problem is if there is slack in your plan to reduce spending. One problem I see is if you don't have margins in your plan for the unknown.

Sadly, I feel that those who are lower income may have less slack. Your SS is lower even though it is more generous for lower income.

 

[pacergal]

Our pension plan had a calculator you could run once a year prior to retirement that would give an estimate of your pension amount. You had to work until age 58 or 30 years to get full amount.
I would run that occasionally for information.
We also saved in a 457b 5-10% of income.

We had planned on working until 65, but around age 50 or so, I started running every retirement calculator we could find.
I also tracked spending for years, so I had that information to compare.

 

[pb4uski]

I found the best retirement planner is Quicken Lifetime Planner which is included in Quicken Deluxe and higher. As you go through the various input screens it make you think of things that you might overlook. The only problem with that too is that it is determiistic and doesn't test for sequence-of-returns-risk, so after I had my retirement pan dialed in on Quicken Lifetime Planner I ran my situation through FIRECalc and other free stochastic planners to assess sequence-of-returns-risk. but if your ultimate WR after pensions and SS are being received is less than 4% then you're probably fine from a sequence-of-returns perspective.

For SS timing I used when I planned to start SS, which in my case was 70. While we debate it incessantly (and I'm guilty as well) the reality is that if you have a well fnded plan then 62 vs FRA/67 vs 70 doesn't make a huge difference in safe spending.

A longer retirement isn't going to change thins much. If you put $1,000,000 in a 60/40 portfolio in FIRECalc with a 30 year time horizon and solve to safe spending at 95% success you will get a WR of 4.06% which is 25x and consistent with the 4% rule/Trinity Study, which is premised on a 30 year time horizon. Change that to a 40 year time horizon and safe withdrawals are 3.68% which is 27x and 50 years is 3.38% which is 30x.

For expenses I used our actual expenses with adjustments for things that would change in retirement but I think just as good an approach is to solve for safe spending using FIRECalc's Investigate tab and then ask yourself can I live the retirement lifestyle that I desire on this amount annually?

 

[frank05]

Also include any costs on how you plan to keep busy in retirement. We decided to travel in luxury for as many years possible. Friends brought a larger boat and spend several months cruising.

 

[jollystomper]

The more you know what you are "retiring to", the better you can estimate your future expenses. I looked back at 5 and 10 year periods to estimate the past expenses that would carry forward, and then factored in thew future expenses we wanted to spend on.

To avoid under estimating expenses, consider taking the "worst case" scenario. For example, for health insurance premiums from age 60 to 65, I looked at the ACA numbers (since we were not eligible for subsidies). I then added an annual growth factor greater than inflation. This exercise was part of my overall "building a buffer" for expected expenses. I also looked at a "worse case" of taking SS at 63, to determine a 3-5 year cash buffer to cover expenses beyond by pension. I ran the results through several models (FIRECalc, Fidelity Retirement Income Planner, model from a personal finance company provided as a Meacorp benefit) to see what lifestyle we could expect.

My initial plan was for 5 years, with an annual review to see how I was doing and if any changes might be needed within the buffer (e.g. whether or not I needed to take SS).
The next result was that my plan overestimated our expenses (especially health insurance, for several reasons). We have spend the way we wanted and are still "underspending" what the models say we can comfortably spent, and I have not yet taking my SS. I am fine with that .

 

[Yegey]

Another question is when fidelity starts to matter. If you are in your 20's, it doesn't seem all that realistic to create a detail plan. There are too much unknown. When do you think is close enough to start to do a more detail plan?

 

[JackJester]

I have been a member on this forum for 10 years. There are many folks that post their questions about FIRE (as did I) and ask whether they are ready or on track for FIRE. But the # 1 item that most don't have prepared for is to clearly understand their current expenses. One needs a spreadsheet that has a categorized list of their current monthly expenses (i.e., gas to groceries to restaurants to utilities to home/health insurance to property/fed/state/ taxes, etc etc etc). I have 30 different expense category items that I tracked before FIRE, and continue to track after FIRE. If you don't know what you need to live on "comfortably" now, then you can't determine whether your savings and investments can cover your comfortable living over x years into the future.

 

[Diogenes]

If one is psychologically ready for retirement, then a decently resourceful person will make the numbers work. If one is unready for retirement, then even a billionaire who eats out of trash cans, won't be able to make the numbers work.

My own "retirement number" is when I will have reached a sense that my career has either properly concluded, or I'm unable to continue it. FireCalc is silent on that.

 

[Koolau]

I knew I was surviving on my salary at w*rk. I used the 4% rule and realized I could take quite a bit more than my salary from my stash. That was then my "number" I guess you would say. When I found this Forum, I revised everything a bit, using FIRECalc. But everything was fine as it was. I always figured I could cut back if need be at some down time.

Best luck. If you have specific questions, be certain to ask.

 

[TheWizard]

I have been a member on this forum for 10 years. There are many folks that post their questions about FIRE (as did I) and ask whether they are ready or on track for FIRE. But the # 1 item that most don't have prepared for is to clearly understand their current expenses. One needs a spreadsheet that has a categorized list of their current monthly expenses (i.e., gas to groceries to restaurants to utilities to home/health insurance to property/fed/state/ taxes, etc etc etc). I have 30 different expense category items that I tracked before FIRE, and continue to track after FIRE. If you don't know what you need to live on "comfortably" now, then you can't determine whether your savings and investments can cover your comfortable living over x years into the future.

Sort of.
While I did put together a spreadsheet of my expenses before cutting loose, it turned out not to be that useful.

What I eventually ended up doing was arranging income streams hitting my checking account each month of roughly the same amount of $$$ as my net income when working.
This exceeds my basic expenses by a considerable amount and has worked out well...

 

[Dtail]

I have been a member on this forum for 10 years. There are many folks that post their questions about FIRE (as did I) and ask whether they are ready or on track for FIRE. But the # 1 item that most don't have prepared for is to clearly understand their current expenses. One needs a spreadsheet that has a categorized list of their current monthly expenses (i.e., gas to groceries to restaurants to utilities to home/health insurance to property/fed/state/ taxes, etc etc etc). I have 30 different expense category items that I tracked before FIRE, and continue to track after FIRE. If you don't know what you need to live on "comfortably" now, then you can't determine whether your savings and investments can cover your comfortable living over x years into the future.

Same here except I use 16 categories, but also a few sub categories.

 

[old medic]

Note, when I said estimate of expense, it is actual but in reality expense may not be the same at retirement so it is at best an estimate. Hopefully it won't change much.

I guess we are lucky in the fact we managed paycheck to paycheck for so many years, both worked (DW still working till Oct) to a fair pension. We figured out our est pensions and spent several years just taking that amount and the rest went to 401Ks.

 

[marko]

Agree with everything above, most importantly, understanding actual expenses.

I was thrust into an unplanned retirement and, at the time didn't have the benefit of this forum or FireCalc, or anything, for that matter.

I just looked at my gross pay and knew that, after removing our monthly saving, it was more than enough to cover our living. Then I just figured out if our portfolio could generate that amount of income... it did.

Caveat: the company had been paying our health care insurance. Now I had to come up with an extra $26k+ for 13 years.

 

[big-papa]

I am curious the method that one would arrive at the retirement number. My thought were

1. Estimate your current expense and probably predict any future creep.
2. Get an estimate of SS.
3. Subtract your SS and other retirement income (like a pension) from the expense.]
4. Estimate 25x for a 30 years retirement.
5. Run some calculation using different methods to check how realistic the plan is. I think typical recommendation is about 90%.
   • A direct method using a pessimistic, and average.
   • A historical; calc like firecalc or ********.
   • A Monte Carlo like in portfolio Visualizer or Fidelity Retirement Planner.

However, things never go as plan. I am curious about the following.

1. How do you avoid underestimating expense? My thought was that expenses must be recalculated every couple of years.
2. How do you handle that SS estimate will vary depending on when you file? I have known people that are forced to retire early and had to file early. My current mitigation is to be able to at least survive with no SS.
3. 25x expense is a good number for a 30 year retirement. What would be the expense for a longer or shorter retirement?
4. Do people even find the calculator useful? They seems to be very input sensitive. Calc result often vary a lot because they have different assumptions on inflation rate, tax treatment, etc. I suppose investigating their difference can dig up assumptions that you can see if you need to incorporate into your model.

Not everyone wants to be this detail. What's the simplest plan you can think of?

1. Yes, but work with today's numbers and do all of the math in "real" dollars (meaning today's dollars) instead of nominal which would require an estimate of inflation.
2. Yes (again that will be in today's dollars)
3. Yes, if you're planning on starting retirement exactly when you start SS. But if you're not, then see #4
4. No because 25X is based on the (so-called) 4% rule (1/4%=25) which is really only a 4% back-of-the-napkin idea of what might have happened if you retired in the past. But you're not retiring in the past and the future has no obligation to follow the past. Instead, at least for quick calculations, I would suggest starting your "funded ratio". Mike Piper lays it out nicely in this article

What’s Your Funded Ratio? — Oblivious Investor

obliviousinvestor.com

And it includes a link to an online spreadsheet here: https://view.officeapps.live.com/op/view.aspx?src=https://obliviousinvestor.com/FundedRatio.xlsx&wdOrigin=BROWSELINK
5. Maybe - Fidelity retirement planner is OK, but you need to spend the time understanding the data set it uses and how it may crudely map your actual holdings onto the data it actually uses. To me, it's was just another tool but I would never have relied on it for my retirement.

Once you're comfortable with that, I would suggest moving on to VPW over on bogleheads. I'm not a VPW user as I have my own spreadsheet, but you can use it the way I used my own spreadsheet. Enter your numbers as if today is the day you're retiring. Stress test it using the feature it has to simulate what sort of drop in an annual withdrawal you would have with a large single year stock drop, and whether you could tolerate that.

Alternatively, you can give the same information with some expected future dollars and start to guesstimate how long it will take you to get there (you need to assume some sort of inflation adjusted returns from between now and when you expect to retire and update the dollar amount in the spreadsheet accordingly)

Regarding the second set of questions
1. We never budgeted and never tracked our expenses at any level of detail. So before I retired, I did a tops/down view of expenses starting with my paycheck and a year's worth of outflows from our checking account, removing any "lumpy" items, reimbursements for travel at work, money moved into taxable brokerage account. etc. On the paycheck side, I started with gross income and removed anything that was no longer relevant to retirement (SS withholdings, Medicare withholdings, 401K and HSA withholdings, etc). Healthcare premiums were removed and then replaced with a best guess ACA plan premiums. This plus a tax estimator like dinkytown gave us what I think we needed to see what our baseline spending would be for a year. We treat discretionary spending differently than the sporadic, lumpy expenses (new roof, new car, etc, unexpected repairs, vacations, etc.). That got us started and was used as the input to what I describe upstream to let us know when we could retire.
2. We used opensocialsecurity.com (also from Mike Piper) to estimate our starting dates for SS. Note that it has a lot of knobs and the main one we changed was to use one of the lifespan tables that we believe more accurately reflects us (eg non-smoker preferred). It came back with wife starting at 66 and me at 70. And that's what we're following. Knowing that, we know the dollar amounts (again in today's dollars) that we'll receive and when.
3. In my opinion, No. See above.
4. Start with calculating your funded (sometimes called funding) ratio for a better estimate of readiness. Then work your way up to VPW. Now both of these are spreadsheets, but both of them only require that you enter data into them in the designated cells. You don't have to be a spreadsheet whiz to do that and you can do all of it in googlesheets for free.

Cheers.

 

[Yegey]

Just to elaborate. I used to budget using quicken when I was younger, but at some point I abandon it when I got married (also quicken started to become a subscription) and instead use the pay yourself first method. Now I am getting closer to retirement, I am starting to try to track my expenses again. Part of it is that I need to separate out kids expenses.

One problem with tracking my expense is that my 20 year old self is nothing like today. There is life style creep over the decades, but that is mostly because my 20 year old self lived in a bombed out apartment in a dangerous neighborhood and that wasn't desirable in the long run. At that point in time, my feeling is that I need to concentrate on saving and investing and staying invested rather than coming up with numbers.

My parents never budgeted and seems to naturally spend less when less money comes in. They have problems getting used to retirement income due to not arriving as a paycheck and end up underspending. According to one of the post on Rational Minder Forum, the average withdraw rate for senior across all level was around 2.8%. If you are used to sparsity, this does not change when you retire and you are unlikely to run out of money but it does not help you come up with the number and you still need to invest. Many retiree may save enough but failed to invest.

I am trying to get some sort of number together. I do have one in my head that is in the ballpark, but I figure I better double-check that it's correct.

 

[pjigar]

Retirement planning can be simple or complicated. But the most important ingredient is the annual expenses. The expenses must include prorated portion of lumpy expenses (roof, car, furness, etc.), taxes, healthcare, etc. SWR of anywhere from 2%-5% may be used depending on the retirement period and asset allocation. Speaking of which, the next most important variable is your asset allocation. e.g. if you keep all your money in CD or cash, your SWR may be dismally low. Either way, you should backtest your number using historic calculator like Firecalc. Firecalc can't tell you how the future will be but typically history rhyme.

As to the "number", it will keep changing every year based on changes to your expenses (which will go up due to inflation/lifestyle changes) and your portfolio (which will go up due to savings and growth). The changing number is very normal. FWIW our "number" has more than doubled since I started retirement planning over two decades ago. I always say that the number is not that important but your funded ratio aka WR is more important for retirement planning.