FIRE on 4%

[pb4uski]

See my post above yours.

Ah... I see.

 

[ggbutcher]

Which brings me to my final point: you'll never get a 100% success ratio, and statistics indicates that anything over 80% is ludicrous. Instead of maximizing your success ratio eliminate your failure rate by annuitizing a portion of your portfolio to provide a minimal standard of living. Maybe that annuity is SS, or maybe it's another type of deferred annuity, or maybe you buy a SPIA. But once you have that minimum longevity insurance covered, then you can invest in a much higher asset allocation of around 80/20 stocks/cash.

And here is what I think is just about the most useful point made on any retirement forum I've read to date.

 

[scrabbler1]

When I was putting together my ER plan back in 2007-08 (well before I first found this forum), I had never heard of WR. Instead, I was working on predicting my ER expenses and my reliable ER income starting at age 45, the age I would be at the end of 2008. For me, even if I had heard of WR, it wouldn't have been too useful for me in my plan because I would have unfettered access to only a portion of my portfolio until I turned ~60/. SS, my frozen company pension, and clean access to my IRA (what I call my "reinforcements") were still at least 15 years away so I could use only my current, non-retirement assets. Should I use a higher WR based on only 2/3 of my portfolio even if I'd be using only that subset for 15 years?

Beinf a Fido client, I met with my Account Executive in 2008 and we ran their RIP program on my data and she gave me a clear signal that I was "clear to go" with my ER. The cute chart showed me with huge surpluses starting at age ~60 after a decent 15-year period prior to age ~60 when I'd be living only off the non-retirement stuff.

Since I ERed, my WR using my entire portfolio has been around 2.5% except for a few years when it was under 2% (some lower HI premiums prior to the ACA). If I excluded the IRA from the denominator, I'd have a WR between 3.5% and 4% but as I wrote above, how meaningful is that?

A good, rebounding market has helped keep the WR down but the bond fund whose income I rely on has seen a slow but gradual decline in monthly dividends since 2009 although I have more shares in that fund now to offset the cents-per-share dividend it has been paying. And I still turn a surplus on average every month to amply cover any small, unforeseen expenses. Isn't that what really counts in the end?

 

[jkern]

I like your guideline. Just wondering are you also considering an inflation adjustment with those figures?

Yes, I assumed that the SWR is the initial rate and adjusted for inflation there after.

 

[hankster]

Great points!
One of the things that I'm slowly grasping is that while I am just starting out in retirement now, (I'm 62, DW is 58), every few years things are going to change for the better. in 3 years I get Medicare, in 4 years DW gets SS, in 7 years DW gets Medicare, in 10 years a 10K/ year mortgage gets paid off... each one of those milestones takes a good chunk off of what I need to withdraw.

With 16 months until planned ER date, I've been encouraged by this same realization. Once we complete the 2 1/2 year gap between retirement at age 59.5 and SS eligibility at age 62, we can look forward to a similar series of income increases and expense decreases over the following 5 or so years. It definitely increases my confidence level to pull the plug next July.

 

[daylatedollarshort]

We looked at what income we would have between SS, pensions and a 2% real return (changed to 1% now that low rates look like they are here to stay for some time) and adjusted our budget to live off that amount or less, ages 65+. Before age 65 we've been living off pensions, investments, and hobby income, though I don't count the hobby income in the retirement plan so we don't have to work unless we want to.

 

[FreeBear]

Which brings me to the next point: we don't rigidly spend 4% + CPI every year. I don't think anybody does, and we can all cut back during a recession. There's another margin to let a portfolio recover from a bear market.

Which brings me to my final point: you'll never get a 100% success ratio, and statistics indicates that anything over 80% is ludicrous. Instead of maximizing your success ratio eliminate your failure rate by annuitizing a portion of your portfolio to provide a minimal standard of living. Maybe that annuity is SS, or maybe it's another type of deferred annuity, or maybe you buy a SPIA. But once you have that minimum longevity insurance covered, then you can invest in a much higher asset allocation of around 80/20 stocks/cash.

+1

Good to hear the voice of reason and real world experience! I agree about the excessive preoccupation with "100% success" ratios on various simulators and calculations. It's likely that most folks on FIRE or anywhere near it have undertaken at least a few business/w*rk projects and life endeavors that were far less than 100% certain to succeed, at least when guesstimated at the start. After all, if it were easy and certain everyone would be successful in everything!

I've spent my life taking calculated risks, and a few dumb un-calculated ones. So far, so good

Some examples:

Guesstimates of odds at beginning or middle of major "projects":

Find love of life and not getting divorced by middle age: <<25%
Well, I married late, but this may have reduced the divorce odds. Still married to the love of my life for 7 years. Married improved with FIRE

Odds of keeping job amid mini-Mega outsourcing and offshoring: <50%
Survived 2 decades, then FIRED when I had enough. Was one of the in-house offshoring experts for the last decade...

Odds of success of my typical mini-Mega development project: about 50%
We were our own worst enemy. Still, batted above .500 over 2 decades of cleaning up other people's problems.

Odds of retiring before 50 yrs old: maybe 50-70%
My ignorance about what it would take to get to FIRE OVERESTIMATED the time and $$ needed (poor understanding of my needed expenses). Still FIRED at 46 yo.

In big contrast to the above, I'm 100% certain that my DW and I will die one day. So we know we must live well with whatever we have left.

We've rolled with the punches before, we'll manage to make it work from day to day.

FB

 

[AdrianC]

Apology accepted... I didn't think that was your intent but I guess I was in a sour mood when I read it.

And to be clear, while we have early retired and our WR is currently more than 4%, once my pension and our SS are online our WR will be much lower than 4% which is why I was comfortable retiring early. For situations like ours I think it most prudent to look at the ultimate WR rate which can be estimated by reducing the withdrawals numerator by SS and pensions and reducing the resources denominator by the amount of funds needed to carry you from ER to the point that SS and pensions start. What you are in effect doing is segregating the money you need to carry you from ER to SS from the total and then calculating your WR at the time you start SS.

Thanks, no problem, and yes, that all makes sense.

 

[FreeBear]

I agree completely that having money as security yields more happiness than just spending it because you can.

To me the 4% rule comes into play when all income streams come online. I plan on starting SS at 62 and that's when I'll evaluate my initial withdrawal rate. Of course I've made some projections, but I won't know what my portfolio value will be until then.

The 4% is a guideline and here's my view on it:

4% Too high, not comfortable at all
3.5% Maximum amount, still worried
3% Comfortable, just ok
2.5% Very comfortable, no problems
2% Absolutely golden

Interesting, we're about 5 years from early SS but are targeting and comfortable with "only" a 3.5% long term SWR (after SS), 40 yr planned withdrawal. I guess the main point is that everyone has to establish a comfort level based on their specific circumstances.

Key factors for us:

- Don't have to leave a big estate when gone

- No current dependents (no minor kids, no deadbeat kids, no living parents)

- Very poor family health on both sides emphasizes importance of living well the next 10-15 years instead of the last 10 years.

- Willing to be more frugal in retirement than when wage slave. We bought a bunch of nice crap when w*rking. It's still serving us well as old nice crap. Surprisingly lost most of the taste for long expensive vacations upon FIRE. Love our little house, hobbies and friends/church. Traveling has become somewhat of an expensive pain.

FB

 

[AdrianC]

Well, I think the reason is that your logic and your conclusion are flawed. The reason is that many early retirees are tired of rehashing this perpetual topic and didn't bother to respond.

You're probably correct. Thanks for your replies.

Luckily if you're not comfortable with the 4% SWR then you're the only one who has to work longer to pad your nest egg until you can sleep comfortably at night. Behavioral finance is at least as important as the math.

Sure. We're good.

Let's rehash the assumptions that the Trinity trio made to simplify their computer simulations:

I don't worry about the Trinity study. We have better tools now (******** and FIREcalc). My plan comes out well on both with > 90% success rate.

Which brings me to the next point: we don't rigidly spend 4% + CPI every year. I don't think anybody does, and we can all cut back during a recession. There's another margin to let a portfolio recover from a bear market.

Exactly. We will be the same.

My plan is a variable withdrawal rate ranging from 2.5% (current base expenses) to 4% (charity, big vacations, big ticket items, sports cars, motorcycles, etc). The lower rate when in capital preservation mode in down markets. The higher rate when we're feeling flush. I'm also continuing to work part time, for this year at least. No extra savings required. Social security for me in 18 years, spouse in 21 years, $20K pension for spouse in 19 years. I assume a 25% cut to SS in my calcs, just to be safer.

I will think about the annuity idea. I kinda think of some of our equity investments that way.

Thanks again.

 

[dallas27]

Well, I think the reason is that your logic and your conclusion are flawed. The reason is that many early retirees are tired of rehashing this perpetual topic and didn't bother to respond.



Luckily if you're not comfortable with the 4% SWR then you're the only one who has to work longer to pad your nest egg until you can sleep comfortably at night. Behavioral finance is at least as important as the math.



But I've been retired for nearly 14 years on the 4% SWR, including two awe-inspiring recessions, and I think that counts as "so far so good" for sequence of returns risk.



Let's rehash the assumptions that the Trinity trio made to simplify their computer simulations:

- 1% expense ratio on investments

- No Social Security

- No flexible spending

- Conservative asset allocation



So if you go by the 4% SWR, right away you're assuming that you throw away 100 basis points every year. In my case, with my expense ratio of about 24 basis points, I have an extra 0.76% on my side.



I don't know about you, but I expect Social Security to be available when I turn 70 years old in 15(!??!) years. That's about $12K/year for me and another $12K for my spouse. That's at least a quarter of our spending unless we're really blowing it out for travel.



Which brings me to the next point: we don't rigidly spend 4% + CPI every year. I don't think anybody does, and we can all cut back during a recession. There's another margin to let a portfolio recover from a bear market.



Which brings me to my final point: you'll never get a 100% success ratio, and statistics indicates that anything over 80% is ludicrous. Instead of maximizing your success ratio eliminate your failure rate by annuitizing a portion of your portfolio to provide a minimal standard of living. Maybe that annuity is SS, or maybe it's another type of deferred annuity, or maybe you buy a SPIA. But once you have that minimum longevity insurance covered, then you can invest in a much higher asset allocation of around 80/20 stocks/cash.



By the time I'm 70 I'll have lived through 29 of the 30 years. I'll let you know how it goes while I reset the calendar for a second 30-year retirement. I'm guessing it'll work out too.

+ a whole bunch


I would jump on this pile with the fact that the study actually concluded a number slightly higher than 4% that the public rounded down for headlines. I think it was really like 4.53% or something like that. I'm sure a former engineer or analyst around here will precisely correct me.


Sent from my iPhone using Early Retirement Forum

 

[liberty53]

Let's rehash the assumptions that the Trinity trio made to simplify their computer simulations:
- 1% expense ratio on investments
- No Social Security
- No flexible spending
- Conservative asset allocation

Nords,

Can you confirm that the Trinity study included the 1% expense ratio? I thought both Bengen and the Trinity study ignored taxes and transaction costs. I can't find any reference to the 1% in the 1998 paper by Cooley, Hubbard, and Walz.

 

[audreyh1]

I would jump on this pile with the fact that the study actually concluded a number slightly higher than 4% that the public rounded down for headlines. I think it was really like 4.53% or something like that. I'm sure a former engineer or analyst around here will precisely correct me.

I think that was the Bengen paper that was published a year before.

 

[Sycamore]

A lot of great responses to an interesting thread.

Recently I read an article by Scott Burns where he referenced a 2006 study published in the Journal of Financial Planning by Jonathan Guyton and William Klinger.

The study: "Decision Rules and Maximum Initial Withdrawal Rates," profiles portfolio withdrawal rate decision rules which generate a greater than 4 percent withdrawal rate for a 40 year time frame.

Here's a link to the Burns article: https://assetbuilder.com/knowledge-center/articles/how-we-live…-and-how-long-our-money-lasts

 

[papadad111]

Which brings me to my final point: you'll never get a 100% success ratio, and statistics indicates that anything over 80% is ludicrous. Instead of maximizing your success ratio eliminate your failure rate by annuitizing a portion of your portfolio to provide a minimal standard of living. Maybe that annuity is SS, or maybe it's another type of deferred annuity, or maybe you buy a SPIA. But once you have that minimum longevity insurance covered, then you can invest in a much higher asset allocation of around 80/20 stocks/cash.

And here is what I think is just about the most useful point made on any retirement forum I've read to date.

There is a flawed argument. SPIA's are quite dangerous in periods of higher inflation. Anyone not thinking inflation will eventually return over the course of the next 30-40 years is naive.

Emerging markets and millennial demographics possibly to be the driver of higher inflation along with continued central bank easy money.

Sure we may be a decade out but we could have periods of inflation not totally unlike the 1970s as driven by the boomer demographic over the next 30-40 years.

Those SPIAs won't be looking so good then. The whole annuity idea is ok if those annuities are cola'd but most SPIAs and deferred annuities are not.

 

[pb4uski]

Valid point but don't forget the 80% in stocks under that strategy... those stocks will provide the growth to overcome inflation.

You have to look at the whole... not each part.

 

[NW-Bound]

... Link to the Trinity Study paper http://www.aaii.com/journal/article...inable?utm_source=sitesearch&utm_medium=click

Thanks for posting the link. I first learned on this forum about the 4% WR and that it was the result of this paper. I never thought of searching for the original paper and read it myself. Now there is no excuse to not read it, but I have learned so much since I just came to this forum that I do not expect to see any surprising data from this original article.

 

[pb4uski]

+1 Thanks for posting the link. I think I had read it before. Some things that caught my eye given debates we have here:

Most retirees would likely benefit from allocating at least 50% to common stocks

For retirees with significant fixed costs and for those who tend to spend less as they age, CPI-adjustments will likely cause a suboptimal exchange of present consumption for future consumption.

For stock-dominated portfolios, withdrawal rates of 3% and 4% represent exceedingly conservative behavior. At these rates, retirees who wish to bequeath large estates to their heirs will likely be successful. Ironically, even those retirees who adopt higher withdrawal rates and who have little or no desire to leave large estates may end up doing so if they act reasonably prudent in protecting themselves from prematurely exhausting their portfolio.

 

[Nords]

I didn't know that the Trinity Study assumed a 1% expense ratio or other cost on investments?

I was pretty sure that they assumed no investment expenses/expense ratios but just used historical index data. Obviously not quite real world. OK - you made me go back and scan it (link at bottom).

And that it was a much later Pfau model that assumed 1% expense ratio among other factors that resulted in a much lower SWR.

Good point, Audrey, my mistake. I confused the expense ratio of Pfau's study with the earlier studies.

 

[papadad111]

Valid point but don't forget the 80% in stocks under that strategy... those stocks will provide the growth to overcome inflation.

You have to look at the whole... not each part.

To be sure one must look at the whole picture.

My point is that a SPIA setup today that delivers the "minimum to survive" will in periods of inflation quickly fall "below the minimum" and therefore one must rely on remaining equities position to fund the difference. However assuming 80/20 that implies 40 percent of total portfolio jn equities - and that may not be adequate to get one "back up to their minimum" without significant principle spend down of those equity assets...

Then one is left with a SPIA not covering the basics AND a seriously depleted stock portfolio (also no longer adequate for covering basics).

Scary

 

[GravitySucks]

+ a whole bunch


I would jump on this pile with the fact that the study actually concluded a number slightly higher than 4% that the public rounded down for headlines. I think it was really like 4.53% or something like that. I'm sure a former engineer or analyst around here will precisely correct me.


Sent from my iPhone using Early Retirement Forum

+1 I started this journey by reading "A Random Walk Down Wall St." And the 95% SWR was said to be 4.5x%

to the OPs post...
I met with my 401K Fido rep and he suggested I look into the 4% rule and then we ran RIP. RIP and fFIRECalc both give a good to go ( for me that's over 90% success to a point that I only have a 10% chance of still needing money.) with a well over 6% <b>initial</b> WR. I haven't exceeded that number yet, but last year was well over 4% and I'm not worrying about it.

I am counting on SS later on, have another year left on the mortgage, just finished helping DD through trade school, and also see my family members fitting that lowered Berneke spending rate after 70. All those factors let me sleep at night even with an outrageous 5+% initial WR.

 

[FreeBear]

I am counting on SS later on, have another year left on the mortgage, just finished helping DD through trade school, and also see my family members fitting that lowered Berneke spending rate after 70. All those factors let me sleep at night even with an outrageous 5+% initial WR.

We started ER last year with a 4.5% to 5% SWR. We've got about 5 yrs before early SS, which would pull the SWR to 4%, probably even less, with a 40 yr planning horizon. We regularly look for ways to cut expenses that would have minimal impact on fun. Now that we are free, we don't seem to need as many expensive "bribes and bling" to get through each day

Understand regarding Berneke, but I'm too much of a worry wart. I expect certain expenses to decrease, and we are even seeing this only a year into ER. Yet, like many folks I'm concerned about healthscare costs, etc.

I would be interested in hearing others' experiences who VOLUNTARILY reducing spending with age due to changing interests/ability instead of running out or fear of running out of $$$.

 

[FreeBear]

Those SPIAs won't be looking so good then. The whole annuity idea is ok if those annuities are cola'd but most SPIAs and deferred annuities are not.

Just did some very casual looking online, looked like those cola'd SPIAs were very pricey at this point in time. As much drag as having a typical assets under management FA, perhaps 2% a year. That's half or more of my budget for the next 4 decades, I'd have to go back to w*rk...

Without cola, I'm losing 2-3% yearly to inflation, plus still some drag from the SPIA. Can't afford this stuff, so resorting to DIY with low cost indexing, 50/50 (or 60/40, etc) AA with yearly rebalance, and hanging on for the ride...

Some suggest SPIA for only a smaller part of the egg nest, but I can't stomach the costs. There's always early SS in 5 yrs if needed/desired, we'll see what the market brings/takes!

FB

 

[tmm99]

I have been reading this thread with great interest. I do look at the rate of withdrawal in relation to my total portfolio, but like some of you mentioned here, it matters if you have SS, pensions, etc in addition to the total disposable asset.

I ran my numbers against VG planner, FIDO RIP, orp and of course, firecalc before taking the final leap. My projected annual spending was about 20K to 28K lower than the amount they said I could spend, so that's how I determined that I may be OK to retire now. My current WR is definitely less than 4% (more like 2.5% - much less than the 3.5% I had in mind, due to USD being so strong against CAD), so even without SS kicking in later on (I should get about $35K SS at age 70), I will most likely be OK, is what I am hoping for.

With this in mind, I don't know why I wouldn't spend more money. I should be able to like some of you are doing. I may start spending more in a few years, but for now, it's hard enough that I have to spend the money I had been so diligently saving (with no more income coming in!) - that I feel this is as much as I am willing to spend for now without much fear.

 

[audreyh1]

Thanks for posting the link. I first learned on this forum about the 4% WR and that it was the result of this paper. I never thought of searching for the original paper and read it myself. Now there is no excuse to not read it, but I have learned so much since I just came to this forum that I do not expect to see any surprising data from this original article.

It's actually pretty basic - in the sense that there are just a few scenarios modeled.

Bengens paper(s) seem more complete. Runs more scenarios, looks at longer times, finer granularity of allocations.

But I heard about the Trinity Study way before I found out about Bengen which was published the year before, I believe. In fact - I had stumbled onto Firecalc before finally reading the Bengen paper.

There has been a lot of work done since.

I read somewhere that numbers had been run through 2009, and the Trinity Study conclusion still stood. It might have been Pfau who did this. If so - that was before he pealed off to explore more depressing scenarios.