6% a year seems pretty fool proof, any thoughts?

[pb4uski]

I think OP's 6% is different than the typical 3-4% SWR.

The SWR gets adjusted for inflation each year... I believe OP's plan (from what I've read) is to take 0.5% per month (6% annually), each month... no adjustment (always taking the 0.5%/6%) except his nest egg increasing as his assets hopefully appreciate over time. This means his annual increase (or decrease) would be dependent on market performance and not inflation.

I don't think there's a direct correlation between the two values (I could be mistaken and I haven't read every post in the thread).

Actually, I forgot that you can model this in FIRECalc. Below is a scenario of 100% stocks, 6%/year withdrawals ($60k on $1000k base).

A constant spending model is designed to allow you to maintain your lifestyle throughout the term of the plan, but variable spending plans use calculated spending amounts that may or may not be suitable after decades. Here is the annual spending for all 117 of the 30 year cycles, shown in 2017 dollars.

From the graph it looks like there are nunerous scenarios where the retiree woudl have to reduce their spending/standard of living with a 6% of balance approach.

FIRECalc looked at the 117 possible 30 year periods in the available data, starting with a portfolio of $1,000,000 and spending your specified amounts each year thereafter.

Here is how your portfolio would have fared in each of the 117 cycles. The lowest and highest portfolio balance at the end of your retirement was $364,743 to $2,477,022, with an average at the end of $1,042,716. (Note: this is looking at all the possible periods; values are in terms of the dollars as of the beginning of the retirement period for each cycle.)

In other models in FIRECalc, "failure" means the portfolio drops to zero. Since you are limiting spending to a percentage of your remaining portfolio, the total balance should never reach zero — but it could become pretty small in some situations. Pay attention to the spending graph, below. Since we can't use portfolio failure as a metric, FIRECalc is following the lead of the 95% Rule from Work Less, Live More, in which one of the goals is for the portfolio to be as big (after adjustment for inflation) at the end of the 30 years as it was when you started. FIRECalc found that 47.0% of the time, the portfolio you would have left behind exceeded the portfolio you started with.

I think this means that 53% of the time that a 6% of balance WR would not allow your spending to keep pace with inflation and possibly required lower spending/lowered standard of living.

 

[medved]

I understand that various SWR analyses can be useful as guidelines, but does anyone really withdraw and spend X% of his/her portfolio every month, or year, no matter what...?

I assume that, in the real world, people spend various amounts, over time, depending on what they want or need to buy in a particular month or year.

And I assume that most people probably adjust their spending based (in part) on how their portfolio is doing.

Actually, I don't have any idea what "most people" do. But I would guess they do these things, and I think I will do these things when I retire.

 

[braumeister]

I assume that, in the real world, people spend various amounts, over time, depending on what they want or need to buy in a particular month or year..

That's what I do.

However, I also track it, with an eye toward how the WR is going.
So far it has been within the 4% guideline (except for one year) and more usually within 3%. As long as it stays in that range, I'll keep spending freely.

 

[ESRwannabe]

pb4uski,

Firecalc, Bogleheads, Trinity, Fama French small/value, etc. give a false impression that future returns can be calculated. This is wrong. Past returns have no more determination on future returns than do past results of flipping a coin determine the future result of a coin toss.

You cannot claim 4% will work nor that 6% will not work in the future. This model only works for the past few decades in the US. Take it to Japan or wherever and the equivalent Trinity study fails.

If we are going to use a limited amount of US data, then it is equally valid to go back and cherry pick anything we want. Someone could do a study on a portfolio of soft drink companies and even make a firecalc for it.

It can become so popular that eventually it attracts a devoted following that goes around calling themselves Cokeheads, and telling everyone on every single finance discussion board that they have the answer, and anything different is wrong. Anyone that disagrees will be told about how firecoke says this strategy allows an X% withdrawal rate with a 95.99% confidence. If you disagree your wrong because just look, we have 40 years of data for the US and it worked just fine there.

 

[NW-Bound]

If we drew 6% of present value each year, then after 10 years the worst-case ending balance of a $1M portfolio was the following, according to FIRECalc.

100% equity: $340K
75% equity: $384K
50% equity: $391K

Note that in the worst case, the $60K/yr WR became $340K * 0.06 = $20.4K/yr. I think the retiree would have hit SS long before the 10-year period ended.

The fact is that in the past, a $1M portfolio without any WR on it demonstrated an amazing ability to shrink occasionally. Just by making some simple runs with FIRECalc, I see that in the worst case, after 10 years of 0% WR, a $1M portfolio became

100% equity: $625K
75% equity: $695K
50% equity: $715K

 

[braumeister]

Firecalc, Bogleheads, Trinity, Fama French small/value, etc. give a false impression that future returns can be calculated. This is wrong. Past returns have no more determination on future returns than do past results of flipping a coin determine the future result of a coin toss.

You cannot claim 4% will work nor that 6% will not work in the future. This model only works for the past few decades in the US. Take it to Japan or wherever and the equivalent Trinity study fails.

If we are going to use a limited amount of US data, then it is equally valid to go back and cherry pick anything we want. Someone could do a study on a portfolio of soft drink companies and even make a firecalc for it.

Perhaps it would be good to review what FIRECalc does and does not claim to do.

https://firecalc.com/index.php

How can FIRECalc predict future returns from past performance?
It can't. And it doesn't try. In fact, it tries to predict what will not happen.

 

[audreyh1]

I understand that various SWR analyses can be useful as guidelines, but does anyone really withdraw and spend X% of his/her portfolio every month, or year, no matter what...?

I assume that, in the real world, people spend various amounts, over time, depending on what they want or need to buy in a particular month or year.

And I assume that most people probably adjust their spending based (in part) on how their portfolio is doing.

Actually, I don't have any idea what "most people" do. But I would guess they do these things, and I think I will do these things when I retire.

I withdraw X% every year, no matter what. I don't usually spend it all, and save the excess for rainy days or splurges.

 

[RunningBum]

I understand that various SWR analyses can be useful as guidelines, but does anyone really withdraw and spend X% of his/her portfolio every month, or year, no matter what...?

I assume that, in the real world, people spend various amounts, over time, depending on what they want or need to buy in a particular month or year.

And I assume that most people probably adjust their spending based (in part) on how their portfolio is doing.

Actually, I don't have any idea what "most people" do. But I would guess they do these things, and I think I will do these things when I retire.

I'm pretty sure there are some people here who do exactly that, withdraw a set amount per month or year and live off that.

I withdraw as I need to. I probably check 2-5 times a year to see if my spending seems to be on track to my budget. If I know I'm hitting more big expenses, I'll check more often.

I don't spend more if more portfolio is doing well for the most part, because I don't want to have to cut sharply back during a correction. I say "for the most part" because I use VPW, so after a good year I'll be able to spend a bit more the next year. If my portfolio went up $100K, I might get an extra $3K or so to spend the next year, because I'm basically spreading out that $100K windfall over the rest of my life, and it can get canceled out in the next downturn.

I'm guessing you're taking some issue with the preciseness of these numbers? I think for planning it makes sense. I'd rather have a clear spending/withdrawal target than a fuzzy one. If I'm a bit over one year, that's fine, but if it's year after year, and investment returns aren't great, it's eventually going to catch up to me. I don't want to have to cut back on things that are closer to "needs" than "wants" in my later years because I was lax in my younger years.

 

[medved]

How is this for my simpleton approach:

4% probably works. It has pretty much worked in the past. But there is no guaranty it will work in the future. So if you can spend a lower percentage, and still be happy, that would probably be a good idea. If you need to spend 4% in order to meet your needs and be happy, then maybe just keep an eye on your portfolio and adjust your spending depending on how your portfolio does. A little bit of adjusting in down years might help.

6% might work. But there's a much worse chance that it will work than 4%. So its higher risk. If you don't mind higher risk (more chance you will run out of money), then go for it. But to pretend that 6% WR does not have materially greater risk than 4% WR seems silly to me.

 

[audreyh1]

I have modeled 6% of remaining portfolio for the case of 50% total stock market, and 5 year treasuries. It survives indefinitely of course, but the income in real terms has dropped 73% in the worst case scenario during a 30 year period. Now, if you are ignoring inflation it might not be obvious in nominal terms, but I expect someone would notice their standard of living was dropping.

Average ending portfolio was 60% of starting portfolio and lowest ending portfolio was 30 % of starting portfolio in real terms.

So it's viable, at least in the 50/50 case I modeled, you are just going to have a gradually shrinking income in real terms from such a high draw.

It's easy to model % remaining portfolio scenarios using FIRECALC.

 

[audreyh1]

Some people like to withdraw only out what they need to spend from their long term investments. This should maximize growth over the long term.

Some of us prefer to take withdraw what history/models say we can take out safely (which depends on various criteria we have selected) regardless of whether we actually spend it all. By not reinvesting unspent funds for the long term, they are no longer subject to market volatility and available for spending in the near term. You give up some long term growth potential, but if one's portfolio is already more than adequate, that extra long term growth will just go to heirs.

Different philosophies. It's like the people who have guaranteed income (pensions, SS) already meeting their needs: some choose to invest in 100% equities because they can take the risk, and some invest 100% in fixed income because they don't have to take the risk. Neither is wrong.

 

[audreyh1]

How is this for my simpleton approach:

4% probably works. It has pretty much worked in the past. But there is no guaranty it will work in the future. So if you can spend a lower percentage, and still be happy, that would probably be a good idea. If you need to spend 4% in order to meet your needs and be happy, then maybe just keep an eye on your portfolio and adjust your spending depending on how your portfolio does. A little bit of adjusting in down years might help.

6% might work. But there's a much worse chance that it will work than 4%. So its higher risk. If you don't mind higher risk (more chance you will run out of money), then go for it. But to pretend that 6% WR does not have materially greater risk than 4% WR seems silly to me.

Just understand that the 6% scenario the OP was proposing does not run out of money. So there is not that specific risk involved. It's rather how quickly his portfolio will shrink over time, but it won't go to zero.

 

[medved]

Just understand that the 6% scenario the OP was proposing does not run out of money. So there is not that specific risk involved. It's rather how quickly his portfolio will shrink over time, but it won't go to zero.

That is a fair point. I was thinking it terms of 6% of the original (at retirement) value. The "percentage of remaining assets" approach never runs out of money. But if your portfolio goes down to one dollar, then you can spend only 6 cents that year. So you might be eating cat food.

 

[audreyh1]

That is a fair point. I was thinking it terms of 6% of the original (at retirement) value. The "percentage of remaining assets" approach never runs out of money. But if your portfolio goes down to one dollar, then you can spend only 6 cents that year. So you might be eating cat food.

Going down to $1 doesn't happen, even for 6%. You might be looking at 1/4 of your original income for one year in real terms before recovering in the absolute worst case scenario for 50/50. You are generally looking at gradually shrinking income if you exceed 4.3% in the scenarios I modeled. But you would have had higher income in the first 15 years, so even dropping farther it takes a while before you drop below the income of a lower starting percentage. So there are tradeoff rather than absolutes

It's hard to really grasp the tradeoff without running a lot of models in FIRECALC.

 

[Texas Proud]

Just understand that the 6% scenario the OP was proposing does not run out of money. So there is not that specific risk involved. It's rather how quickly his portfolio will shrink over time, but it won't go to zero.

I remember when I was taking calculus in high school the teacher would say 'the answer is approaching 2 so the answer is 2'.... BUT, it actually never does get to 2... WHAT My strict logical head never did like that thinking...


So sure, it never will be zero, but there IS a possibility that it approaches zero that we can call it zero... if you start with $1 mill, income is $60,000... but if the portfolio drops to $100,000 (not likely, just as an example) then income is now $6,000.... that is close enough for me to say you have run out of money.... and if the portfolio lost 10% then that is only $84,000 for the next year which is now down to a bit more than $5,000....

 

[pj.mask]

The most interesting parts of this thread (to me) were the discussions around historical projections. Not sure if I should worry we're in for doom and gloom or expect things will be fine and dandy. I suspect this is something I will keep in mind over the next 15 years. I would love to say with confidence that growth over the next 15 years will be 5-6% after inflation and that 3.5% will be a SWR that should grow my portfolio over time... with all this talk of uncertainty vs reward - maybe stock will do much better for me since I have so much uncertainty

 

[audreyh1]

I remember when I was taking calculus in high school the teacher would say 'the answer is approaching 2 so the answer is 2'.... BUT, it actually never does get to 2... WHAT My strict logical head never did like that thinking...


So sure, it never will be zero, but there IS a possibility that it approaches zero that we can call it zero... if you start with $1 mill, income is $60,000... but if the portfolio drops to $100,000 (not likely, just as an example) then income is now $6,000.... that is close enough for me to say you have run out of money.... and if the portfolio lost 10% then that is only $84,000 for the next year which is now down to a bit more than $5,000....

I explained that with the historical scenarios it doesn't even get close to zero with 6%. 27% of your starting income real, is nowhere close to zero, and it recovers after that. 30% of starting portfolio in real terms was the worst case ending portfolio after 30 years.

It's tough to say what constitutes "running out of money" in the % remaining portfolio case. If you still have 30% of what you started with in real terms, after 30 years worst case, and you don't have hardly any years left, and 60% of what you started with real on average, it's hard to call that running out of money.

You can pick your rate based on how much belt tightening you are willing to live with and/or the lowest or average ending portfolio. But ironically, if you are starting out with a larger %, your income is a lot higher, so your lowest income ends up not being much lower than a more modest rate. For example, with 6% withdrawal $1M portfolio, your lowest income would have been $16,248, but with a 4% rate your lowest income would have been $17,419, just a little higher. In the first case your starting income was $60K, and in the second, $40K. So even though you took out 50% more at the start, and your portfolio suffered a much more drastic reduction in the worst case historical scenario, you just had a slightly lower income than the 4% case.

% remaining is not a good method unless you have a lot of discretionary expenses and have come up with ways to deal with income variability, because under the rare worst case historical scenarios you would have to deal with serious belt tightening - 66% drop for the 4% case and 49% drop even for the very modest 3% case.

 

[medved]

I explained that with the historical scenarios it doesn't even get close to zero with 6%..

OK - but "with historical scenarios" is not the same thing as "absolute worst case."

Really, it all depends how risk averse you are. Most people, i think, are sufficiently risk averse that they are not willing to consistently spend 6% of their portfolio. Whether the risk they are guarding against is "nearly running out of money" or instead just "having less money than enables me to have what I regard as a decent quality of life" its a risk most are not prepared to take.

But, hey, its a free country. If someone wants to take that risk, they should have at it.

 

[audreyh1]

OK - but "with historical scenarios" is not the same thing as "absolute worst case."

Really, it all depends how risk averse you are. Most people, i think, are sufficiently risk averse that they are not willing to consistently spend 6% of their portfolio. Whether the risk they are guarding against is "nearly running out of money" or instead just "having less money than enables me to have what I regard as a decent quality of life" its a risk most are not prepared to take.

But, hey, its a free country. If someone wants to take that risk, they should have at it.

You can't quantify that "absolute worst case" for ANY method. All we have is historical data.

People have to run the models and make their choices. It's better that they understand how various scenarios compare, rather than make assumptions. FIRECALC is great for that.

I don't want to leave a large amount of $ behind. I would rather spend and gift it during my lifetime. So I consider under withdrawal to be as serious a problem as over withdrawal, so I look for balance.

 

[dallas27]

my plan is to take 0.5%per month which works out to 6% per year with a 100 % equity allocation. This plan is guaranteed to never run out of money. Our base expenses are low enough that could easily survive a 50% drop in the stock market. Also we could always start collecting social security if the market drops more than 50% We are 60 years old and are eligible for about 66k in ss payments per year at age 70 which we could survive on worst case.

It seems pretty simple to me so I must be missing something. Please give me your thoughts

You should read a lot more than you have thus far, and here is a good place to do so.

 

[Sunset]

I explained that with the historical scenarios it doesn't even get close to zero with 6%. 27% of your starting income real, is nowhere close to zero, and it recovers after that. 30% of starting portfolio in real terms was the worst case ending portfolio after 30 years.

.....

I have no idea how firecalc calculates or what it considers, but purely from a historical perspective, the market has shown it can go lower.
Facts Stock Market Crash 1929 | Facts The Great Depression

"September 1929, the Dow Jones more than doubled, increasing from 191 points to 381 points .... By July 1932, years after the start of the Great Depression, the Dow Jones was at just 41 points."

So the market ended up at only 10.8 % of it's high value.

 

[pb4uski]

pb4uski,

Firecalc, Bogleheads, Trinity, Fama French small/value, etc. give a false impression that future returns can be calculated. This is wrong. Past returns have no more determination on future returns than do past results of flipping a coin determine the future result of a coin toss.

You cannot claim 4% will work nor that 6% will not work in the future. This model only works for the past few decades in the US. Take it to Japan or wherever and the equivalent Trinity study fails.

If we are going to use a limited amount of US data, then it is equally valid to go back and cherry pick anything we want. Someone could do a study on a portfolio of soft drink companies and even make a firecalc for it.

It can become so popular that eventually it attracts a devoted following that goes around calling themselves Cokeheads, and telling everyone on every single finance discussion board that they have the answer, and anything different is wrong. Anyone that disagrees will be told about how firecoke says this strategy allows an X% withdrawal rate with a 95.99% confidence. If you disagree your wrong because just look, we have 40 years of data for the US and it worked just fine there.

You obviously don't understand how FIRECalc or some of these other studies work. They don't suggest that future returns can be calculated so they certainly can't give a false impression that future returns can be calculated.

While I concede that past returns are not necessarily indicative of future results, what are you suggesting that people use? A ouija board? Or just trust the obvious wisdom of some wannabe and ignore the past?

Given the basis for conclusion that you have provided why isn't it 10% or 20% rather than 6%?

And actually, FIRECalc has 145 years of data (since 1871), not just 40 but please... don't let facts get in the way of your arguments.

I'll take my chances with FIRECalc and other such tools.. good luck with your 6% WR, if history is any indication, you're going to need all the luck you can get.

 

[pb4uski]

I have no idea how firecalc calculates or what it considers, but purely from a historical perspective, the market has shown it can go lower.
Facts Stock Market Crash 1929 | Facts The Great Depression

"September 1929, the Dow Jones more than doubled, increasing from 191 points to 381 points .... By July 1932, years after the start of the Great Depression, the Dow Jones was at just 41 points."

So the market ended up at only 10.8 % of it's high value.

So you are chosing to ignore the safeguards put in place since 1929 to specifically prevent rampant speculation?

 

[teejayevans]

I have no idea how firecalc calculates or what it considers, but purely from a historical perspective, the market has shown it can go lower.
Facts Stock Market Crash 1929 | Facts The Great Depression

"September 1929, the Dow Jones more than doubled, increasing from 191 points to 381 points .... By July 1932, years after the start of the Great Depression, the Dow Jones was at just 41 points."

So the market ended up at only 10.8 % of it's high value.

The Dow Jones is not the market. This why we diversify.

 

[ERD50]

I have no idea how firecalc calculates or what it considers, but purely from a historical perspective, the market has shown it can go lower.
Facts Stock Market Crash 1929 | Facts The Great Depression

"September 1929, the Dow Jones more than doubled, increasing from 191 points to 381 points .... By July 1932, years after the start of the Great Depression, the Dow Jones was at just 41 points."

So the market ended up at only 10.8 % of it's high value.

Check those (apparently superficial) numbers.

Do they include dividends? Does it include the deflation from that time? When you do that, the 1929 market doesn't look as bad. Like others mentioned, we can't have that kind of bubble again, no 10% margin allowed.

Instead, we get bubbles in the housing market, where 10% margin was considered conservative. We will never learn.

-ERD50