A more realistic way to calculate for your retirement savings

[Gumby]

I hate to think that this guy spent hours and hours putting together and testing a model that may/may not be more/less accurate than the dozen other calculators immediately available online for free.

Besides, there's always a certain margin of error in all these calculators, so I'd wonder if it's better/worse by only 0.0002% or something.

Building your own model is the best way to really understand what drives the results, so that you can take specific action to improve those results. It is also helpful where your personal situation may differ from other, more generic models. In any event, I don't think it is a waste of time to learn how these things work.

 

[walkinwood]

Use your model by all means, but be aware of risks that it doesn't model - like the sequence of return risk. Firecalc does a good job of modeling that risk using historical returns or (IIRC) monte-carlo simulations.

There are a number of threads here discussing the possibility that expenses will decline with age. Take a look at those threads too since that pushes your "number" lower.

Welcome.

 

[target2019]

My calculation finally showed up again at #10.

Your id reveals your intended ER year as 2019. We are in the same boat, except you joined this site years earlier than me. Hopefully, there will be no catastrophic market correction in coming years.

I will check it out.

I actually got there four years early. Megacorp cut me mid 2014, and I am just doing temp work here and there as it shows up. Both feet are in the tub and it feels really good.

 

[marko]

Building your own model is the best way to really understand what drives the results, so that you can take specific action to improve those results. It is also helpful where your personal situation may differ from other, more generic models. In any event, I don't think it is a waste of time to learn how these things work.

Early on, I actually did build my own model. Because I'm not as smart as those who created FireCalc and others, I omitted a few key considerations which could have had potentially disastrous results.

In my case anyway, if you're not smart enough to NOT need a model, maybe you're not smart enough to build your own.

 

[itsme2]

I'm not sure how you arrived at the bold statement.

Maybe a numeric example would help. Suppose we try a very simple retirement. A single person, with

a: Annual expense for the first year: $40,000

c: SS of what you can withdraw:$18,000.
n1: The number of years before you start to withdraw SS: 0
n2: The number of years after you begin to withdraw SS: 30

You can specify r1, r2, and r3, plug them into your formula, and get a value for b.

Then, you can specify what number "most people here" would use for b, and show why you think it is "too conservative".

Here is the result based on the numbers given above.

a $40,000 Input
b $461,771 Output
SS $18,000
r1 1.06
r2 1.03
r3 1.02
n1 0
n2 30


How does $461k look like?

 

[itsme2]

Building your own model is the best way to really understand what drives the results, so that you can take specific action to improve those results. It is also helpful where your personal situation may differ from other, more generic models. In any event, I don't think it is a waste of time to learn how these things work.

That is exactly correct. I initially did this on a spread sheet based on my own situation (That is no consideration of pension, because all our pensions with previous employers were converted to 401k). As I played with the numbers again and again, I thought it would be fun to convert a spreadsheet results to a more generalized formular.

There are lots of existing guesstimates available. However, I believe this is the first one that makes the "magic" lump sum directly depend up your annual expense. If you need $a, you will know exactly how much $b should be.

 

[brewer12345]

There are lots of existing guesstimates available. However, I believe this is the first one that makes the "magic" lump sum directly depend up your annual expense. If you need $a, you will know exactly how much $b should be.

So the trinity study that says you need 25X your expenses isn't magic enough for you?

 

[gerntz]

I don't get the obsession to know the answer to within a gnat's hair. The effort to try to achieve that level of knowledge isn't productive to enjoying life imo. In fact, I think it's a waste. Get/make a good estimate & live on 90% of that & adjust every few years. What difference is that initial 10% going to make to the next 30 years?

 

[itsme2]

What is your basis for believing most people here are "too conservative?"

And for anyone who might be interested in your approach, what probability of success does your formular offer? It's relatively simple to project the average portfolio needed to support a given spending for a given number of years, assuming a real rate of return - but that may give you a 50% probability of success. Good enough?

You and several others questioned why I said existing models are “too conservative”. First of all, it is purely subjective, as different people have different risk tolerance levels. While the average soon-to-retire people have relatively much lower retirement savings balance, people thinking about ER and joining this site, including myself, are generally “conservative” with our retirement planning approach.

My model is still conservative. 1). by assuming that SS COLA is behind the inflation rate; 2). by assuming that the investment return rate is lower than the generally accepted 8% of historical US stock market return.

Models are just for planning and do not guarantee anything. Many people here use FireCalc and are looking at “success rate” in the end. My questions for you are: 1). what success rate is considered high enough and can be accepted? 2). unless you can achieve a success rate of 100%, have you thought about “what-if” bad luck do come?

To me, for retirement planning, the annual expense is a number for comfortable living, not for bare minimum. If the market goes down (especially in earlier retirement years), we can certainly reduce our living standard. It is no different from those who saw their retirement savings tanked due to market meltdown and decided to delay their retirement. Bad thing will occur, but it is never the end of the world.

I am aware that many people seek peace of mind by accumulating a higher figure. But, in the meantime, they are spending extra years on the job, that otherwise may not be necessary. So, it is a balance of choice.

 

[itsme2]

So the trinity study that says you need 25X your expenses isn't magic enough for you?

Like what I said earlier, those factors (25X or whatever) are irresponsible. It doesn’t take your available SS in to account. Also, it doesn’t tell you whether you can retire at 40, 60 or 70. The retirement age and your available SS, along with inflation & investment return rate, do significantly affect the outcome.

By how much? My model delivers the answer.

 

[RobLJ]

I tend to agree, with huge caveats, that many here are probably too conservative. As I have noted before, I intend on using a 5-5.%% withdrawal rate after DW retires or semiretires, until SS, when we will probably reduce. Or not. It will depend on sequence of returns the next 7 years.
Caveats: our "needs" withdrawal rate is about 2.5%, with only a small mortgage but with 3x taxable savings. Why not pay out the mortgage? Because of the concerns above about making it to SS.
We can cut back travel, entertainment, and other categories, and health insurance is only 250/month. The less slack in the budget the more the sequence of returns will kill you, if you can't adjust (no 10-20k Europe trip).
I'm dialing back to half time work, which I should be able to expend for a while. DW is 5 years younger and wants to continue to work, for a while.
So we have huge flexibility with 1-1.5 partial incomes.

You and several others questioned why I said existing models are “too conservative”. First of all, it is purely subjective, as different people have different risk tolerance levels. While the average soon-to-retire people have relatively much lower retirement savings balance, people thinking about ER and joining this site, including myself, are generally “conservative” with our retirement planning approach.

My model is still conservative. 1). by assuming that SS COLA is behind the inflation rate; 2). by assuming that the investment return rate is lower than the generally accepted 8% of historical US stock market return.

Models are just for planning and do not guarantee anything. Many people here use FireCalc and are looking at “success rate” in the end. My questions for you are: 1). what success rate is considered high enough and can be accepted? 2). unless you can achieve a success rate of 100%, have you thought about “what-if” bad luck do come?

To me, for retirement planning, the annual expense is a number for comfortable living, not for bare minimum. If the market goes down (especially in earlier retirement years), we can certainly reduce our living standard. It is no different from those who saw their retirement savings tanked due to market meltdown and decided to delay their retirement. Bad thing will occur, but it is never the end of the world.

I am aware that many people seek peace of mind by accumulating a higher figure. But, in the meantime, they are spending extra years on the job, that otherwise may not be necessary. So, it is a balance of choice.

 

[brewer12345]

Like what I said earlier, those factors (25X or whatever) are irresponsible. It doesn’t take your available SS in to account. Also, it doesn’t tell you whether you can retire at 40, 60 or 70. The retirement age and your available SS, along with inflation & investment return rate, do significantly affect the outcome.

By how much? My model delivers the answer.

Except you are ignoring perhaps the biggest risk for early retirees: sequence of returns risk. Your model is frankly quite simplistic.

To illustrate the sequence of returns risk issue, I think this is a pretty good explanation: John Hancock - GIFL Solutions - Sequence of Returns Risk

Note that the hypothetical early retiree in the historical example they use who ran out of money was actually the one who had higher returns over time.

I know you will say "that does not account for social security." Straw man if I ever saw one. You can account for SS in firecalc while getting the benefit of modelling sequence of returns risk. In any case, ask 10 random people under 40 if they think they will be getting SS...

 

[Midpack]

Models are just for planning and do not guarantee anything. Many people here use FireCalc and are looking at “success rate” in the end. My questions for you are: 1). what success rate is considered high enough and can be accepted? 2). unless you can achieve a success rate of 100%, have you thought about “what-if” bad luck do come?

We all use whatever success rates we're comfortable with, typically from 75-200% - no one is under the illusion there's a narrow range of "right" answers. If we hadn't thought about the bad luck "what if's", we'd all use more optimistic success rates.

To me, for retirement planning, the annual expense is a number for comfortable living, not for bare minimum. If the market goes down (especially in earlier retirement years), we can certainly reduce our living standard. It is no different from those who saw their retirement savings tanked due to market meltdown and decided to delay their retirement. Bad thing will occur, but it is never the end of the world.

Sure it's easy to adjust early on, that's not the big question. The challenge is in projecting the range of likely outcomes 20-40 years on. Are you going to go back to work in your 80's or 90's when it becomes obvious your plan is going to fail? Might be too late when you realize it, could be "the end of the world" as you've known it during your first 20 years retired. Most here plan "too conservative" to avoid living in a box under a bridge eating cat food, it's a fair exchange for many here.

I am aware that many people seek peace of mind by accumulating a higher figure. But, in the meantime, they are spending extra years on the job, that otherwise may not be necessary. So, it is a balance of choice.

Yep, a choice we all have to make, and live with - despite what anyone else may think. Most here would rather be safe than sorry, but there is no way you can know if you worked extra years/saved more than needed/spent less until the END of plan - no model can tell you that in advance, not FIRECALC or any other. I am not seeing how your simple model is more helpful, except as your self educating tool.

 

[karluk]

My model is still conservative. 1). by assuming that SS COLA is behind the inflation rate; 2). by assuming that the investment return rate is lower than the generally accepted 8% of historical US stock market return.

This is one of the most peculiar methods of being conservative that I've ever heard of. Essentially, you are making multiple mistakes in your inputs that you hope will offset each other and lead to a more "realistic" retirement estimate. You have no way to include sequence of returns risk in your model, so you intentionally low-ball SS COLAs, in spite of the fact that by law SS COLAs are determined by the inflation rate. Sheesh, you may call your model "realistic" , but I would tend to call it GIGO. Still, if it works for you, go for it. Just don't expect me to follow your lead and base my planning on such a flawed model.

 

[imoldernu]

We didn't start planning with "what we would need", but from what we had.
From an old post, this is how we've handled our savings. (The only change from this older plan, is that we now look at age 90)

There are hundreds of financial planners on line where you put in your estimates of assets, and return and inflation, and come up with the amount you need to retire. In our case it doesn't work... All of the planners make the assumption that you will want to maintain your asset capital until you die... In our case, had we followed their plan, we NEVER would have retired.
We just decided to die at age 85... dead broke. Made our planning much easier. Personal decision of course, but if you plan to spend down capital assets, it makes planning easier.

Our plan is extremely simple... On the spending side, we have three different budgets that we can adjust as circumstances warrant. Best case... Nominal... and Austerity.

On the Asset/Nest Egg side, We boil our assets down into three categories.
1. Fixed assets... house, auto, and other valuable non cash items... real property, jewelry, . We do not count household goods... (experience tells us that this is not realistic)
2. Non Income producing assets... bank accounts, cash, cash value life insurance policies.
3. Income producing assets... stocks, bonds, annuity.

All of these items are kept on a spread sheet and periodically updated. It's easy to come up with a total value... and then to average the income from the total...

To calculate where we stand in our retirement plan, we add
a. Social security amount.
b. Amount of interest earned on income producing assets.
c. ... and add the Total Assets divided by the number of years between now and age 85.

That establishes how much we can spend, which we then adjust to our best/nominal/austerity budget.

Sounds funky, but it works,and it takes about 2 minutes to tell if we're on budget or not.

The second part of this budgeting thing, is that we've been blessed by not having any debt. All of this makes for very simple accounting.

Not for everyone, but has worked for us for the past 25 years.

 

[Rosie]

I'm still very happy with Firecalc. It lets me work out different scenarios and it's very easy to use.

As a token of my appreciation, I just sent off a very small donation via the paypal link (I've done this a couple of other times over the past 5 years or so).

 

[bingybear]

Firecalc and most of the planning programs are quite conservative when you consider success at 90% to 100%. The reason it is conservative is that you are looking at the cases that closely fail. So for many users... they will have better results that those that nearly fail. The 90-100% are on the tails of the distribution. Many will end up in the middle of the distribution... and will have lots more to spend. This is a statistical analysis as one does not know what the returns...or the sequence of returns are. Firecalc uses historical data vs random data for monte carlo calculators. Itsme2... you are picking fixed returns. So who is more accurate? realistic?

Instead of using a 95% for firecalc... what if you use 60% for success?
The real key is to understand what the calculations mean... yours or firecalcs. They are estimates of SWR.

 

[Golden sunsets]

Imoldernu; what happens when you are 86(or 91 now that you have revised that portion of the formula). If you are drawing down your assets as you head toward your age of death (originally 85), can you suddenly revise your computation to 90 and still have enough of an asset base to generate sufficient annual income? Why did you choose such low dates of death? It seems very risky.

 

[itsme2]

Imoldernu; what happens when you are 86(or 91 now that you have revised that portion of the formula). If you are drawing down your assets as you head toward your age of death (originally 85), can you suddenly revise your computation to 90 and still have enough of an asset base to generate sufficient annual income? Why did you choose such low dates of death? It seems very risky.

Planning is not a guarantee as we are all smart enough to know. No one will use any model to do his estimate and stick to that number for the next 20-40 years. Market can be up and down. You health condition can continue to be good or get worse as time goes by. That is why you can do the calculation every year and adjust accordingly.

Typical calculation can be done with an assumed life expectancy. As you get older, even the lift expectancy gets extended. For example, you can look at how retirement accounts' RMD is calculated, as one ages.

 

[rodi]

itsme2 -

I agree that there are some folks on this board who are VERY conservative in determining their "number" before retiring. But it's all in the details - which vary from person to person.

Some folks here want to cover their expected budget in dividend returns - and never touch the principal... That's fine, but it requires a very large nest egg.

Others want to leave a large legacy to their children. That will also require a bigger nestegg.

That said - plenty of people here are comfortable spending down to zero or near zero ... but would rather not go negative (run out of money).

I used firecalc, ********, financial engines, i-orp, my own spreadsheet, fidelity RIP and every other calculator I could get my hands on. Some gave me a smaller nest egg number than others... But they were only shooting for a 95% success rate.

For me, personally, I wanted a 100% with firecalc - plus left the option on the table to cut expenses and/or sell my primary home (not included in the nest egg) if things go south.

Others are fine with an 80% success rate in firecalc... Everyone has different levels of comfort and what will let them sleep at night.

Your spreadsheet isn't a bad thing. Nor is posting here about it to learn what other people have questioned/looked at when developing their own analyisis.

We all have different details. Some folks here have nice pensions, so less need of a nest egg. Others (like myself) have a rental income stream which allows me to have a smaller nest egg. Others are the kings and queens of frugal living - which lets them get by with a smaller nest egg. And there are plenty of people here who have enough money to not worry about this - and can spend freely and not worry. This board is pretty diverse.... I wouldn't say we all retired with huge nest eggs.

 

[Independent]

Here is the result based on the numbers given above.

a $40,000 Input
b $461,771 Output
SS $18,000
r1 1.06
r2 1.03
r3 1.02
n1 0
n2 30


How does $461k look like?

If I plug $461k into FireCalc, with 100% Equities (I think that's your plan), I get a historic 78% "success" rate.

As you said, some people will find that "too many" failures. It appears that you don't.

Are you sure that they are "too conservative"? or maybe just "more risk averse than I am" would be better wording.

You've mentioned that you're willing to adjust spending downward if "times are bad".

Flexibility is definitely a big deal in terms of initial spending. But, you haven't specified the circumstances where you'll adjust downward. What if the market is flat instead of returning 3% real in the very first year, is that enough to make you adjust? If so, how much? What if it's down __%, what will you do then?

Other people have tried to think about those questions. The most common approach is some sort of "percent of current portfolio" spending rule. Here's one that generated nearly 300 comments recently. http://www.early-retirement.org/for...wal-calculator-ive-seen-to-date-68770-15.html
Maybe it works for you. Maybe you'll find that everyone here isn't as conservative as you assume.

Or you could test a percent of current portfolio with a floor. FireCalc is already coded to test that option (see the "spending models" tab).

 

[imoldernu]

What itsme2 said.

Planning is not a guarantee as we are all smart enough to know. No one will use any model to do his estimate and stick to that number for the next 20-40 years. Market can be up and down. You health condition can continue to be good or get worse as time goes by. That is why you can do the calculation every year and adjust accordingly.

Typical calculation can be done with an assumed life expectancy. As you get older, even the lift expectancy gets extended. For example, you can look at how retirement accounts' RMD is calculated, as one ages.

We began planning in 1989...
When I was born, my life expectancy was 59.9 years.
When I retired @ 53 My life expectancy was 71.7 years
Today my life expectancy is 6.8 more years or age 86.
If I make it to 86, I get 5.4 more years.

So far, the actual is ahead of the plan, and if I should make it to 92, I won't give a damn.

If I had listened to the advisor that I hired in 1989, I wouldn't have retired until the year 2000.

My theory is, "If you're happy with what you do, don't worry about retirement... If you really want to retire early, you'll find a way."

 

[rodi]

Firecalc also has a "spending method" that you can input the 95% rule... In other words you drop your spending to 95% of the previous year's spending if there are bad returns.

With the given inputs it gives 100% success - but a whole lot of the spending lines are in the 23k-ish range.... If you *need* 40k - 23k is going to be hard to get by on.

 

[itsme2]

If I plug $461k into FireCalc, with 100% Equities (I think that's your plan), I get a historic 78% "success" rate.

Are you sure that they are "too conservative"? or maybe just "more risk averse than I am" would be better wording.

You've mentioned that you're willing to adjust spending downward if "times are bad".

Flexibility is definitely a big deal in terms of initial spending. But, you haven't specified the circumstances where you'll adjust downward. What if the market is flat instead of returning 3% real in the very first year, is that enough to make you adjust? If so, how much? What if it's down __%, what will you do then?

Is 78% success rate good enough? I have no idea. Because I don’t know how its calculation came about. That is why I asked “what the acceptable success rate should be” earlier in this thread.

Does FireCalc generate the number based on historical market index? If it does, does it include dividend? For example, your stock could go from $50 to $55 in one year (a nice 10% return), but it may also pay out $2 dividend. So your effective return is $7/$50 = 14% (instead of 10%, if calculated purely based on the stock price). The discrepancy will definitely get larger over 2-3 decades.

I would be more comfortable with my $461k result than most people here, since I derived the formular on my own and I tested with lots of different scenarios.

No matter what model we use, when “times are bad”, we will all have to face the reality and cut spending. Even for those who really have sufficient amount of savings, they are likely to spend less due to psychological reason. So, it just won’t be me using my model.

You brought up the criteria of downward adjustment in case of market downturn. I have not put much thought to it, as I have been mainly focusing on improving my model on the initial lump sum. But, in my recent reading, I did find this interesting article, on downward/upward adjustment rule.


Rethinking a 4% Withdrawal Rate

"The initial withdrawal rate is between 4.8% and 6%, based on the stock/bond mix. At the end of each year, the retiree takes the preceding year’s withdrawal amount [in dollars] and adjusts it for inflation. But there are guardrails against big market swings: If that amount divided by the current portfolio balance equals a withdrawal rate of 20% more or less than the initial rate, the retiree adjusts the amount they withdraw that year. No annual withdrawal is more than 10% more or less than the year before." (My interpretation is that, with +/- 20% market swing, you adjust by +/- 10%)

 

[pb4uski]

FWIW, similar to you, I use a deterministic model for my basic planning. While I have an Excel model, I principally rely on Quicken Lifetime Planner and suggest that you check it out.

The shortcoming of QLP and other deterministic models like your spreadsheet is that they fail to account for sequence of returns risk, so I supplement my deterministic planning with stochastic planners like firecalc, ********, Vanguard's Monte Carlo model, Financial Engines and others to "stress-test" my deterministic plan.

What a 78% success rate means is that 78% of the trials did not run out of money, which means that 22% of the time it did run out of money. Whether 78% is good enough is a personal decision, but personally i would shoot for 90% or more. Some people here are not comfortable with anything less than 100% (or more).

https://retirementplans.vanguard.com/VGApp/pe/pubeducation/calculators/RetirementNestEggCalc.jsf is a good tool to start with. If I plug in $460k, $22k initial withdrawal ($40k spending less $18k SS, a 4.8% withdrawal rate), a 30 year time horizon and 100% equities it suggests a 80-82% success rate.