Stop worrying about the 4% rule

[forummm]

A collection of posts about why you should not worry about the 4% rule. Please add others that you find useful.

[forummm]

From Nords:  http://forum.mrmoneymustache.com/post-fire/what-has-workednot-worked-for-you-guys-who-have-been-fire-for-10-yrs/msg700740/#msg700740
(worth reading more in the thread)

Nords - great web site! The Real Life Retire Investment Returns was most interesting to me. The only points which I wasn't clear on in the various portfolios... 1) were all portfolios rebalanced once a yr or just those noted? 2) Was the 4% harvested equally amongst the individual portfolio mixes or was there some sort of order (I.e. Interest, dividends, cash, bonds, last stock sales)?

Thanks! 

I don't think that those differences are significant.  If they were relevant then Greaney would've taken the time to highlight them.

Those questions might not be relevant for any retirees, especially if the advice isn't suited for the retiree.  For example, some are firm believers in rebalancing every 1 January while others just do it when their asset allocation gets too far out of whack (I'm in the latter camp).  Some research says annual rebalancing is better, other research says every 2-3 years.  You have to pick the system with which you're most likely to stick.

You'll harvest your 4% with cash first, then the annual cap gains & dividends & interest, and then you'll sell off the assets that are farthest out of whack from your asset allocation.  So it doesn't matter what the study did-- your real-life practice is pretty close to the study, so minor differences won't be significant.

As a 4 yrs from FIRE guy, here is my takeaway so far for someone who is planning for 40+ yrs RE...
- Understand your risk temperment and don't kid yourself if your a 60/40 guy and going with a 100% stock portfolio

Absolutely.  It's the investor behavioral psychology aspect of investing, where you have to be able to sleep comfortably at night.

- After taking into account any other income sources, 3 - 3.5% may be a safer withdrawal rate

Um, no.  By that logic 0.0% would be even safer.

Consider the assumptions in the 4% SWR studies: 
no annuitized income (not even Social Security),
1% account expense ratios,
constant withdrawals (adjusted for inflation),
no extra cash beyond one year of expenses (just stocks/bonds).

Those are all simplifying assumptions for the benefit of the computer programming.  Nobody lives like that! 

Imagine that some of your income was annuitized (even if it was "just" Social Security).  If you encounter one of the failure scenarios where you blithely spend all of your assets at a 4% SWR, then you'll still have your annuity income.  In other words, your failure rate is actually zero because you have longevity insurance.  Yet the SWR studies never include this reality.

You're going to pay much less in fund expenses.  If you get your expense ratios down to 0.5% instead of 1% then the real SWR becomes 4.5%.  If you're thinking that you'll maintain a 3%-3.5% SWR then just cut your expense ratios and add the savings.  Now you're at 3.5%-4%.

But wait, there's more:  variable spending.  You'll spend more when the markets are up ("wealth effect") and cut back your spending during recessions ("loss aversion").  Some years you'll have a big fantasy vacation or a roof replacement, other years you'll hang out at home most of the year.  During the first five years of ER you'll be particularly sensitive to a recession and you'll probably cut your spending way back when it happens.  Even just using Bob Clyatt's variable spending scheme of 4%/95% in "Work Less, Live More" raises the 4% SWR to about 4.5%.  So variable spending is probably worth another 0.5% added to your more conservative version of the SWR.

Now if you keep two years of expenses in a money market or CD (instead of just one) then your asset allocation is more like 60/36/4.  Yet starting the year with two years' expenses in cash means that you can outlast almost any recession by just spending down your cash (which you're trying to conserve anyway) and letting your stocks recover.

Just by addressing some of those simplifying assumptions, we've taken your 3%-3.5% SWR right back up over 4%.

If you retire on a 4% SWR then you'll have a success ratio of 80%-100%, depending on your personal willingness to work longer to boost it past 80%.  (Anything over 80% is meaningless precision, but it'll help you sleep better at night.)  Your annuitized income (for a bare-bones budget) will guarantee that you'll avoid the failure rates of 0%-20%, so already your actual success ratio is at 100%.  In other words then you'll have more money than you need.

If you cut your spending to 3%-3.5% then you won't be any happier, although you might sleep better at night.  But eventually you'll realize that you have more money than you have health & mobility to enjoy it.  Your great-grandchildren will be happy with your legacy, but you'll have missed a lot of life-enhancing opportunities because you were too conservative with your spending.

Here's another way to look at it:  reducing your SWR from 4% to 3-3.5% is the same effect as cutting your current spending by 12%-25%.  Try that test now.  Cut your spending by that much for the next couple of years, while you're still working, and see how you feel about it.

I think you're good with a 4% SWR and behavioral psychology.  Don't over-think it.

- Using common sense and making sure you have flexibility to lower your annual expenses when needed (market tanks)
- 1 - 2 yr cash/CD/short term bond acct not a bad way to cushion a heavier stock portfolio mix

Yep, as mentioned in my previous paragraphs.

- Be sensitive to stock valuations when you launch and adjust your SWR accordingly if needed

Whatever the heck "valuation" means.  Trailing earnings?  Projected earnings?  EBIDTA?  Free cash flow?  One of the Internet's most notorious personal-finance trolls claims that he's been in cash since 1996 because valuations are still "too high".  Even 2009 wasn't good enough for him because future earnings were projected to be zero, and when a zero is in the denominator of P/E then valuations will always be too high.

A more concrete approach would be to start your ER at your personal conservative asset allocation, say at 60%/40% equities/bonds.  (I would not be more conservative than that.)  If a recession occurs during the first few years of ER then you'll cut your spending.  You'll spend your cash.  You'll spend some of your bonds.  Just by your spending, even though it's reduced, your AA will begin to drift up to 65%/35% or even 70/30.  At the end of the recession you'll resume your 4% SWR, your portfolio will have survived a recession and assumed a more aggressive AA, and it'll recover its value more quickly because you held on to your equities.  You'll have also endured the dreaded "sequence of returns" risk which is critical to portfolio survival during the first 5-10 years of ER.  Once you've survived that then your portfolio is nearly bulletproof.

- Maybe give consideration to a more aggressive SWR in yrs prior to age say 75 with a reduced SWR there after (This is basically my own research/observations as the theory is while you are younger/healthier, you will want to do more that may cost more $$ and as you get in your golden yrs you may be a little less active. Yes, exceptions everywhere, but I see this/hear this from people in my life)

One of the earliest mentions of this was a 1990s book by Michael Stein "The Prosperous Retirement." The anecdotal evidence is strong, but the research is mixed.  Are people spending less because they have to (running out of assets), or because they can't spend it?  (Poor health, limited mobility.)  What about medical expenses or end-of-life care?  In any case, I'm going to agree with the anecdotal data because it closely matches my personal experience.

I think your instincts are correct.  You could start your ER at a 4% SWR, and as you get into your 70s you might find out that your new 30-year SWR is really 5% or even 6%. 

I'm 54 years old, so it'll be interesting to see whether the actuarial analysis reaches a firm conclusion over the next decade.  However I'll still spend a little more now out of concern that I won't be able to spend it later.

[forummm]

From Sol: http://forum.mrmoneymustache.com/investor-alley/bogle-projects-'nominal-to-zero'-real-returns-over-the-next-decade/msg702331/#msg702331
(worth reading more in the thread)

this is why I'm not a fan of FIRE = 25x annual expenses. As

Keep in mind that the 25x rule and 4% rule already account for 95 percent of scenarios, so you would only have a one in 20 chance of needing to ever reduce your expenses.

To extend your hotel analogy, the correct plan would be to take the average expectation, which in this case is a 6% withdrawal rate (has a 50/50 chance of success over 30 years) and then add a 10% safety buffer (another 0.6%) and still plan on a withdrawal rate over 5 percent.  Your plan of going with 40x expenses is like planning for every party that makes a reservation to need the entire hotel, just to be safe, even though that has never happened before in all of history.

Yes, it is safer.  No, it's not a very efficient way to make plans.

[forummm]

From Eric:

Jumping in here, this is why I'm not a fan of FIRE = 25x annual expenses. As brooklynguy stated a few posts back, none of us really know what the future will hold for the market. Instead of skirting by on 25x expenses, I think it's more prudent to go for 30 or 40x expenses. Because if your adaptation alternative is going back to work, you aren't FIRE anymore.

That's totally cool if that's your comfort level, but realize the trade off.  You're talking many extra years of guaranteed work, probably 5-10 years to get to 40x, versus a possible but unlikely part time work here and there if your 25x takes a big hit initially (and you choose to do that instead of implement other safety margins).  So if you go back to work "you aren't FIRE anymore", that still sounds better to a lot of us than never being FIRE to begin with until many years after 25x.

Sol touched on it above, but that 4% is already a pretty worse case scenario.  To add some numbers, the average portfolio balance at the end of 30 years using the 4% rule is over twice of what you started with (in real terms, not nominal).

[forummm]

From Brooklynguy:

To second (third?) Sol's and Eric's point, take a look at this excellent recent post by Nords highlighting the ridiculous amount of safety built into the 4% rule (in that, besides having an excessively high historical success rate to begin with, it deliberately ignores various external levels of safety margin which, in reality, will inure to the benefit of most retirees) -- and he's one of our more conservative early retirement planners!

[forummm]

From Skyrefuge:

If you need to maximize your 25x based on 4%, in the case of any growth oriented portfolio over 60% stock, there needs to be a draw down strategy that plays the law of averages the best.

If you believe in a 4% SWR, then you're totally overthinking all of this.

The research that created the 4% SWR assumed the same withdrawal strategy that forummm is describing: simply withdraw your 4%, and rebalance to your original AA (or withdraw your 4% in a manner that maintains the original AA; that's a mathematically identical way of saying the same thing).

A "down market" doesn't require any different approach. The existence of "down markets" and the simple withdrawal method are baked into the 4% SWR number, and any attempts at fancy withdrawal gymnastics are more likely to hurt your portfolio survival than help.

[a1smith]

Well, let's not make this a totally one-sided discussion.  Here is a post from clifp; I'll also add an old post of mine that discusses a similar concern I have:  While the 4% SWR seems fine for someone that is close to a 30-35 year retirement period it seems to be non-conservative for someone that is trying to retire early in their 30's, for example, and have up to a 60-65 year retirement.

To second (third?) Sol's and Eric's point, take a look at this excellent recent post by Nords highlighting the ridiculous amount of safety built into the 4% rule (in that, besides having an excessively high historical success rate to begin with, it deliberately ignores various external levels of safety margin which, in reality, will inure to the benefit of most retirees) -- and he's one of our more conservative early retirement planners!

I too agree with the sentiment provide your willing to be flexible as noted above and in Nords post.  If not then a lower SWR may be more prudent given the current valuations, lower interest rates, slow growth prospects.

I complete agree about not worrying about the 4% rule if we were talking about people retiring in their 60s.  But according to this calculator http://gosset.wharton.upenn.edu/mortality/perl/CalcForm.html, I as a 55 year old, have a 25% change of living another 34 years (same age as my mom) and if I was a woman a 50% chance.  So anybody planning on retiring before 50 should be looking at 40 year retirement (especially a couple) .  At 40 years a 4% WR only gives 81% success rate according to CFiresim.   Plus life expectancy is increasing at >1.2 year/decade in the US so a 25 year old today is likely to have a life expectancy at age 50 that is 3 years longer than a 50 year old today.   In order to get to a 95% success rate you need to dial the withdrawal rate down to 3.5%.

For somebody looking to retire in their early 40s or before, who has already slashed their lifestyle expenditures  to Mustachian levels, I don't think you are being too conservative to move the SWR rate to 3.0%.  Social security for a 40 year old is 22 years away and conceivably that age maybe raised by a couple of year.

[a1smith]

Here is my old post where I did cFIREsim simulations for longer retirement periods.  In The 4% Rule: The Easy Answer to “How Much Do I Need for Retirement?”, MMM says that "In other words, above 30 years, the length of your retirement barely affects the safe withdrawal rate calculations." (his emphasis, not mine.)  Looking at the cFIREsim calculations you can see that is not the case.  Upthread there is some discussion about 1% fee being high and that you can increase your SWR if you have lower fees -- note that the % success calculations below are for 0% fees.

So, to look at one data point, let's say someone retires at 30 and plans on living till 95.  With 50/50 AA cFIREsim predicts 55.0% chance of success and 80.0% chance of success at 70/30 AA.  So, at the very least, you have to adjust your AA to make sure you have an AA that guarantees a reasonable chance of success but is still an AA you are comfortable with given no income to invest.

Based on the trinity study and cfiresim you should be very safe at 4%. Saying that it only has a 52% success rate shows something is wrong. Also at some point having 50 years or 250 years should have a minimal change to the SWR. I will stick to cfiresim until a better model comes out.

Some food for thought - in MMM's post http://www.mrmoneymustache.com/2012/05/29/how-much-do-i-need-for-retirement/ he shows the 4.04% SAFEMAX value in Figure 2.1.  Actually, that figure is derived from William Bengen's work in 1994 (the author, Wade Pfau, extended the results to 1981 using data up to 2010), not the Trinity study.  The Trinity study occurred four years after Bengen's seminal work.  The only difference between their work is the bond indices they chose.  With Bengen, he showed 100% success at 4.15% WR; the Trinity study showed a 95% success rate at 4% WR (not 100%!!!).

If you go to the webpage MMM got that figure from, it redirects to http://retirementresearcher.com/the-trinity-study-and-portfolio-success-rates/.  Looking at Table 2.1, you can see that at 4% withdrawal rates you have 100% success at 30 years, but only 96% at 35 years and 85% at 40 years.

I tried a really simple experiment in cFIREsim and obtained comparable results.  I started with 2015-2045 for a 30 year retirement, with $100 portfolio, $4 spending (4%) with inflation adjust , 50/50 stocks/bonds, no fees, and everything else set to zero. That matches the study parameters. I then ran simulations for 30, 35, 40, ...., 70, and 75 year retirements. I stopped at 75 years for two reasons: most people won't be retired that long and also the data starts to be statistically insignificant because at 75 years only 70 cycles are ran, 80 years gives 65 cycles, etc.

Well, the results were so bad for 50/50 stock/bond allocation I also ran 70/30.  The results are shown in the table below.  As you can see the success rate keeps dropping and it is not clear to me that there is an asymptotic limit.  I'm not sure where the assumption that if the money lasts 30 years it will last forever came from but I consider it suspect.  If someone can post some hard calculations or a research paper that shows this I would appreciate it.

Years Retired	50/50 Success	70/30 Success
30	90.4%	93.9%
35	82.7%	90.9%
40	67.6%	84.8%
45	65.0%	80.0%
50	63.2%	81.1%
55	60.0%	82.2%
60	60.0%	82.4%
65	55.0%	80.0%
70	53.3%	78.7%
75	50.0%	77.1%

Also, I ran the MMM values in cFIREsim that I used in the spreadsheet.  90/10 allocation, 4% spend, 0.1% fees, inflation adjust, 46 year retirement. I got an 83.84% success rate with cFIREsim.  Higher than the simple spreadsheet but still not 100%.

So, the safety factors MMM mentions in article listed above are not safety factors; they are an absolute requirement.

[forummm]

Good. All viewpoints welcome. This is intended to be a repository of information on the subject. Thanks for contributing!

[MDM]

A related thread is http://forum.mrmoneymustache.com/welcome-to-the-forum/50-60-year-retirement-timeframes-anyone-have-the-numbers/.

There is a comment in the MMM blog post referenced by a1smith: "It’s much like a 30-year mortgage, where almost all of your payment is interest. Drop your payment by just $199 per month, and suddenly you’ve got a thousand-year mortgage that will literally take you 1000 years to pay off."  Unfortunately that is not a generic truth - as the interest rate changes, so does the principal required.  E.g.,

Code: [Select]

Rate/yr Principal
0% $73,900
1% $83,500
2% $98,100
3% $116,000
4% $138,100
5% $165,600
10% $449,800

Much as MMM likely had a specific set of assumptions in mind for the "$199 rule", there are specific assumptions that lead to the "4% Safe Withdrawal Rate rule".  In either case, when different assumptions apply, different conclusions become valid.



ETA:If one wants to skip ~1000 posts, the one below provides a good summary of what is discussed in them:

Here is the live thread: I'm Bill Bengen, and I first proposed the 4% safe withdrawal rate in 1994. Ask me anything!

Highly recommend reading the thread.

His opening remarks and a few intersting points are reposted below.  I hope this is acceptable.

Reposting without comment, as it is of general interest to forum participants:

/financialindependence

I'm Bill Bengen, and I first proposed the 4% safe withdrawal rate in 1994. Ask me anything!

Thanks to ER10years_throwaway for this invite. I was a financial advisor for 25 years, now retired, but still expanding my research into safe withdrawals from retirement portfolios. I am eager to share my thoughts with you, so please bring on the questions. Caveat: I can't answer questions specific to a particular person's financial situation, as I am no longer a practicing financial planner or investment advisor. Hope to hear from you. I'll start answering questions at noon eastern on Tuesday, 8/21.

Q:  Since these questions get asked all the time here:   Is the 4% rule still relevant in today's economy? What safe withdrawal rate would you recommend for someone planning for longer than 30 years of retirement?

A: billbengen • Thanks for your question. Before I answer it specifically, why don't we dispense with some preliminaries, so we are all on the same page?

The "4% rule" is actually the "4.5% rule"- I modified it some years ago on the basis of new research. The 4.5% is the percentage you could "safely" withdraw from a tax-advantaged portfolio (like an IRA, Roth IRA, or 401(k)) the first year of retirement, with the expectation you would live for 30 years in retirement. After the first year, you "throw away" the 4.5% rule and just increase the dollar amount of your withdrawals each year by the prior year's inflation rate. Example: $100,000 in an IRA at retirement. First year withdrawal $4,500. Inflation first year is 10%, so second-year withdrawal would be $4,950. Now, on to your specific question. I find that the state of the "economy" had little bearing on safe withdrawal rates. Two things count: if you encounter a major bear market early in retirement, and/or if you experience high inflation during retirement. Both factors drive the safe withdrawal rate down. My research is based on data about investments and inflation going back to 1926. I test the withdrawal rates for retirement dates beginning on the first day of each quarter, beginning with January 1, 1926. The average safe withdrawal rate for all those 200+ retirees is, believe it or not, 7%! However, if you experience a major bear market early in retirement, as in 1937 or 2000, that drops to 5.25%. Add in heavy inflation, as occurred in the 1970's, and it takes you down to 4.5%. So far, I have not seen any indication that the 4.5% rule will be violated. Both the 2000 and 2007 retirees, who experienced big bear markets early in retirement, appear to be doing OK with 4.5%. However, if we were to encounter a decade or more of high inflation, that might change things. In my opinion, inflation is the retiree's worst enemy. As your "time horizon" increases beyond 30 years, as you might expect, the safe withdrawal rate decreases. For example for 35 years, I calculated 4.3%; for 40 years, 4.2%; and for 45 years, 4.1%. I have a chart listing all these in a book I wrote in 2006, but I know Reddit frowns on self-promotion, so that is the last I will have to say about that. If you plan to live forever, 4% should do it.

Q:   Because your assumption is not correct. The S&P500 is not 'offering' ANY future return. You are likely referring to past returns. Expected future returns are at issue here.

A: billbengen • Expected returns for the S&P 500 over the next decade are, according to a number of sources, very low, possibly zero or less. I recall that between 1966 and 1982, the S&P 500 did return zero. Michael Kitces, a brilliant financial advisor, created a chart matching stock market valuations with subsequent 30-year safe withdrawal rates. The negative correlation is virtually perfect; when stock valuations are high, the safe withdrawal rate was low, and vice versa. His advice, with which I concur, is that when stock market valuations are at very high levels, as they are today, it is best to stick with the appropriate safe withdrawal rate, and not try for something higher.

Q:   Obviously market conditions have changed a lot since 1994. Given that we've been in a bull bond market for so long, and given we're currently looking at corporate bond yields of maybe 4%, does it still make sense to have a bond component in your portfolio?

Also: did you ever foresee the development of a financial independence / early retirement movement like we have today?

A: billbengen • Yes, I still believe bonds should play a significant role in most retirement portfolios. During a stock bear market, interest rates often decline, which causes an increase in the price of bonds. This can offset some of the losses from the stocks. Overall, I believe a 50% equities/50% bonds mixture at the start of retirement is close to ideal. Years ago, I talked to Harry Markowitz, the founder of Modern Portfolio Theory, about this. He used that 50/50 ratio in his personal portfolio, which speaks volumes! Some recent research advocates increasing the fraction of stocks in the portfolio as the retiree ages. I haven't had an opportunity to verify this, but I plan to look into it in the next year. No shortage of intriguing ideas in this field! I think the financial independence movement is great, in part because it means people must educate themselves more in this field so they make good decisions. I have "retired' three times, and am now in my fourth career, as a writer/researcher. But many friends and acquaintances of my generation are still working, even into their late 70's, so I wonder how "early retirement" is succeeding in this environment. Like everything else, if you plan and execute early and well, you will most likely achieve what you want.


Q:  Related to the question on low bond returns, are you bothered at all by the number of people in the FI/RE community who are retiring with 100% stock portfolios?

A: billbengen • It doesn't "bother" me, as when the big stock market decline comes (and it will, eventually), I will not be the one with big losses! All kidding aside, my research indicates that using a 100% stock allocation sharply reduces your SWR. These folks might have to make some major adjustments in lifestyle during a major bear market. But if they are prepared to do so, they might get by.


A few interesting perspectives worth discussing.

Rule revised in Bengen (2006) to 4.5% if tax-free and 4.1% for taxable, underlying assumption, asset allocation of roughly 50/50 equities and debt, with at least 50-55% low cost index funds. The rest in quality bonds and cash (recommends 10% cash).

Q: Inflation assumptions sensitivity

A: billbengen •  It all depends on your view of future inflation. I like to remind people that the 4.5% rule is not a law of nature, like Newton's laws of motion, which will probably never change. Markets can change, and it is possible that in the future the 4.5% rule, which has held up for 50 years, might be violated. But I haven't seen those circumstances yet.

Q:  For a traditional retirement, 4% safe withdrawal rate is a pretty standard assumption for a 30 year period. What is your perspective for using the 4% SWR for a longer period of time, say 60 years? A lot of people pursuing financial independence aim for lower than 4%- it is pretty common to hear SWR's around 3.25%- 3.75% to be more conservative for a longer retirement horizon.

A: billbengen • As your "time horizon" lengthens, my research indicates that you should reduce your withdrawal rate concomitantly. For a 60-year time horizon, the indicated safe withdrawal rate is reduced from 4.5% to 4.0%.

billbengen • It is an interesting fact that in the past, 96% of retirees, at the end of 30 years, have a portfolio still worth at least as much as they started with, in nominal terms. Of course, when inflation is factored in, the real value of those investments has diminished considerably. It should also be noted that the SWR assumes that at the end of 30 years, the retiree will run out of money with his or her dying breath. If you wish to specify a minimum balance at the end of 30 years, that will result in lower initial withdrawal rates.

[waltworks]

I'd just like to point out that the 4% SWR also assumes:
-That you live for 30+ years. You might not.
-That no cataclysmic natural disaster befalls the earth in that time. One might.
-That no social upheaval/revolution/political change makes your earnings and savings valueless.

So, when you say you want a 2% withdrawal rate because it's safer, you're ignoring the fact that it's arguably impossible to be "safer" than, say, 85-90% with any long term plan, because the world may have other plans for you and running out of money in a future that looks just like the present isn't the biggest concern anymore.

-W

[deborah]

The 4% SWR also assumes you live in the USA      and    the USA continues to be similar to what it was in the past.

Other places have much lower SWRs - Australia is about 3.6%, Japan is (from memory) 0.6%. In fact, of all the SWRs I have seen, the one for the USA is the highest. Obviously the hyperinflation of Germany in the 30s (which has a similarly low SWR) and the economic problems that Japan has experienced more recently mean that these countries have much worse SWRs than just about every other country, but we just don't know what is around the corner for our own economies (and as a result for their SWRs) - especially as we are considering the next 50 years.

Throughout the majority of the 20th Century the USA was the largest world economy, so it could be argued that other economies "jerked" twice as much - with their own problems as well as when the global economy (the USA) "jerked". This possibly gave the USA SWR its advantage. However, the Chinese economy (in global terms, not per person) is now, or just about, larger. Over the next 50 years (for the SWR we are postulating), we should consider that other economies (perhaps also India) will probably play leading roles globally, and that the USA economy may become a much smaller component of the global economy. Maybe people should be looking at the SWR of the UK rather than the USA when they are doing their calculations. (The UK "ruled the waves" at the beginning of the 20th century, but gradually decreased over the timeframe used to calculate SWRs).

[arebelspy]

The 4% SWR also assumes you live in the USA      and    the USA continues to be similar to what it was in the past.

Other places have much lower SWRs - Australia is about 3.6%, Japan is (from memory) 0.6%.

No, it doesn't assume it will be similar to the past--it assumes it will be no worse than the past. And those other countries worst case scenarios are worse than ours have been, yes, but I'm curious what their average scenarios were. That would be a lot more interesting to me, but I'm not aware of that data existing.

[Kwill]

I am disappointed. I thought I was going to learn that I could do a 50% withdrawal rate and be OK. Then I could actually retire soon. Too bad. ;)

[Jeremy]

The 4% SWR also assumes you live in the USA      and    the USA continues to be similar to what it was in the past.

Other places have much lower SWRs - Australia is about 3.6%, Japan is (from memory) 0.6%.

No, it doesn't assume it will be similar to the past--it assumes it will be no worse than the past. And those other countries worst case scenarios are worse than ours have been, yes, but I'm curious what their average scenarios were. That would be a lot more interesting to me, but I'm not aware of that data existing.

Even better

It assumes that it won't be worse than the WORST period in the history of the US

As a refresher, that period included an Oil Embargo, the closing of the gold window, discredited economic policies (Laffer Curve anyone?), price fixing, runaway inflation, gasoline shortages, etc...  just about the perfect storm of horrible economic conditions

And yet the 4% rule worked, by definition

[clifp]

Well, the results were so bad for 50/50 stock/bond allocation I also ran 70/30.  The results are shown in the table below.  As you can see the success rate keeps dropping and it is not clear to me that there is an asymptotic limit.  I'm not sure where the assumption that if the money lasts 30 years it will last forever came from but I consider it suspect.  If someone can post some hard calculations or a research paper that shows this I would appreciate it.

Years Retired	50/50 Success	70/30 Success
30	90.4%	93.9%
35	82.7%	90.9%
40	67.6%	84.8%
45	65.0%	80.0%
50	63.2%	81.1%
55	60.0%	82.2%
60	60.0%	82.4%
65	55.0%	80.0%
70	53.3%	78.7%
75	50.0%	77.1%

Also, I ran the MMM values in cFIREsim that I used in the spreadsheet.  90/10 allocation, 4% spend, 0.1% fees, inflation adjust, 46 year retirement. I got an 83.84% success rate with cFIREsim.  Higher than the simple spreadsheet but still not 100%.

So, the safety factors MMM mentions in article listed above are not safety factors; they are an absolute requirement.

I think this table illustrates the limitation of historical calculators.  As you may notice 45 years appears the worse case such that a 60 year retirement is better than a 45.  Both Firecalc and cFIREsim give similar result although the AA matters.

Now if we step back and think about this logically, going forward is is mathematically impossible for a longer retirement to be a safer than a shorter one with the same AA and withdrawal.  A 40 year old who retired in 2000 with a 4% SWR, like I did  is in far worse shape financially than a 65 year. The 65 year now 80 probably only has 10-15 years left to live. The 55 year like myself has 25-35 left. 

Now while it true in a solid majority cases after 30 years you end up with a more and often a lot more money than you start, in a roughly a 1/3 of the cases you end up dipping into your principal, so it is pretty important to understand the limitation of this calculators for early retirements.

I also noticed while playing around with longevity calculators that being retired is probably added a couple of years to my life, much lower stress, more exercise, more sleep.

[deborah]

The 4% SWR also assumes you live in the USA      and    the USA continues to be similar to what it was in the past.

Other places have much lower SWRs - Australia is about 3.6%, Japan is (from memory) 0.6%.

No, it doesn't assume it will be similar to the past--it assumes it will be no worse than the past. And those other countries worst case scenarios are worse than ours have been, yes, but I'm curious what their average scenarios were. That would be a lot more interesting to me, but I'm not aware of that data existing.

Even better

It assumes that it won't be worse than the WORST period in the history of the US

As a refresher, that period included an Oil Embargo, the closing of the gold window, discredited economic policies (Laffer Curve anyone?), price fixing, runaway inflation, gasoline shortages, etc...  just about the perfect storm of horrible economic conditions

And yet the 4% rule worked, by definition

But all these things happened to other economies as well. It is interesting that when those things happened in other countries (for example when run away inflation hit Germany), they DID cause their SWRs to go lower, and the study that talks about why each economy had its worst outcomes (which then feed into the lower SWRs) really shows (to me) that ANY country could have had some of these things happen (perhaps not the things that caused the very low SWRs of Germany and Japan).

[zataks]

Something I started thinking about yesterday and haven't seen anywhere (although "anywhere" means here or reddit) is legitimate arguments against 4% as a SWR.

For almost all theories and beliefs there are valid and legitimate counter beliefs, studies, science.  And I'm just not seeing it regarding this topic.  Is this because I hang out in places where this is essentially the mantra? or is it because nothing has come up to prove otherwise? Or is it because this is relatively simple so the science/maths behind what is going on is not up for debate?

[arebelspy]

Right, so now we're assuming that the future will be worse than anything we've ever had and as bad or worse than any other country has ever had in the last 200 years as well?

Seems pretty paranoid.

If the 4% rule (or 3% for another country, whatever) was based on the average withdrawal allowed to have success (i.e. 50% success rate), I'd share your skepticism, and say the future might be worse than the past (by average).  But to say that the future will be worse than the worst period in our past and all the other countries, too, so one should go sub-3%, or sub-1%, or whatever number you want to argue, seems just way too pessimistic.

I can buy that we'll be worse than average.  But the worst ever?  Seems unlikely. And even if it happened, there's lots you can do in that case.  Seems as good of a reason as any to not worry about the 4% rule.

[arebelspy]

Something I started thinking about yesterday and haven't seen anywhere (although "anywhere" means here or reddit) is legitimate arguments against 4% as a SWR.

For almost all theories and beliefs there are valid and legitimate counter beliefs, studies, science.  And I'm just not seeing it regarding this topic.  Is this because I hang out in places where this is essentially the mantra? or is it because nothing has come up to prove otherwise? Or is it because this is relatively simple so the science/maths behind what is going on is not up for debate?

Huh? There's these arguments all the time.  That's why this thread was started, to compile together the posts arguing why you don't need to worry about it.

[a1smith]

I think this table illustrates the limitation of historical calculators.  As you may notice 45 years appears the worse case such that a 60 year retirement is better than a 45.  Both Firecalc and cFIREsim give similar result although the AA matters.

I just found another nice thread started by sol discussing firecalc and cFIREsim.

firecalc and cFIREsim both lie?

I think it is pertinent to this thread since many people are using those tools to pick a SWR, probably starting with 4%.

[forummm]

I think this table illustrates the limitation of historical calculators.  As you may notice 45 years appears the worse case such that a 60 year retirement is better than a 45.  Both Firecalc and cFIREsim give similar result although the AA matters.

I just found another nice thread started by sol discussing firecalc and cFIREsim.

firecalc and cFIREsim both lie?

I think it is pertinent to this thread since many people are using those tools to pick a SWR, probably starting with 4%.

Yeah, that was another good one. That's where I learned about the difference in the tools. Thanks for adding.

[forummm]

Something I started thinking about yesterday and haven't seen anywhere (although "anywhere" means here or reddit) is legitimate arguments against 4% as a SWR.

For almost all theories and beliefs there are valid and legitimate counter beliefs, studies, science.  And I'm just not seeing it regarding this topic.  Is this because I hang out in places where this is essentially the mantra? or is it because nothing has come up to prove otherwise? Or is it because this is relatively simple so the science/maths behind what is going on is not up for debate?

Some of those arguments are already popping up in this thread. The most common I see is "the SWR is XX% for country YYY". But if you just have a globally diversified portfolio, then it doesn't matter so much if Japan's equity returns sucked for 20+ years, because your eggs aren't all in one country's basket.

[Tyler]

Some of those arguments are already popping up in this thread. The most common I see is "the SWR is XX% for country YYY". But if you just have a globally diversified portfolio, then it doesn't matter so much if Japan's equity returns sucked for 20+ years, because your eggs aren't all in one country's basket.

The 4% rule was not arrived at using a globally diversified portfolio, so this may be a false assumption.

One other thing to remember is that the studies the 4% rule were based on did not settle on stocks and bonds in the US as the best investment options after evaluating everything else.  They studied them at the exclusion of everything else.  Portfolios that use things like international stocks, REITs, TIPS, and other assets may in fact perform better (or may not!), but most retirement calculators do not even offer them as options.  Retirement investing is more diverse and complex than the simple 4% rule of thumb may indicate. 

[zataks]

I guess I should have emphasize legitimate arguments against it? 

So many I see are "4% won't work because of downturns in the economy!" or "4% is not feasible because we can't predict the future!"

I'm curious is the is reasonable, empirically researched data/studies that make a sincere or valid attempt at invalidating a 4% SWR.  In really dumb terms: the Sith to the Trinity Study's Jedi?

[arebelspy]

I guess I should have emphasize legitimate arguments against it? 

So many I see are "4% won't work because of downturns in the economy!" or "4% is not feasible because we can't predict the future!"

I'm curious is the is reasonable, empirically researched data/studies that make a sincere or valid attempt at invalidating a 4% SWR.  In really dumb terms: the Sith to the Trinity Study's Jedi?

I don't understand what you're looking for.  What sort of data would disprove it?  It's a narrow study based on certain premises.  The data supports the conclusion because the conclusion was drawn from the data.

Some attack the premises, or say they should be different, but given the same assumptions, you should draw the same conclusions.  You just may not buy the premises.

[Tyler]

I'm curious is the is reasonable, empirically researched data/studies that make a sincere or valid attempt at invalidating a 4% SWR.  In really dumb terms: the Sith to the Trinity Study's Jedi?

Sure.  By none other than Wade Pfau, the guy who made the 4% rule popular.

http://tinyurl.com/nrywjny

[arebelspy]

By none other than Wade Pfau, the guy who made the 4% rule popular.

Not sure how Pfau made it popular?

And most of his current ideas on why the 4% rule may not work is based on future projected yields.  I don't buy the premise that we can predict future yields, so I don't buy his arguments about why he doesn't trust it going forward.

Pfau writes some awesome retirement stuff, but he's too pessimistic for me.

[Tyler]

I'm happy to walk back the "made it popular" comment.  But there's no denying he's very influential on the topic.

Pfau is probably the most widely quoted guy regarding SWR research, mostly because he continuously updates the original research with new data.  His opinion is definitely more pessimistic lately.  Here's a catalog of his research.  https://ideas.repec.org/e/ppf6.html

I make no judgments on his methodology.  But he's put out some fine research worth noting and is willing to change his views with new information, and with a PhD from Princeton he's no internet hack.

[brooklynguy]

No, it doesn't assume it will be similar to the past--it assumes it will be no worse than the past. And those other countries worst case scenarios are worse than ours have been, yes, but I'm curious what their average scenarios were. That would be a lot more interesting to me, but I'm not aware of that data existing.

Pfau's study from 2010 on SWR's in international markets has some limited data on "average" scenarios.  In particular, Table 3 gives a bunch of data beyond the safemax WR's in various foreign markets, including the failure rates for a 5% WR (which was closer to being an "averagely successful" WR in some of the countries) and it also gives the WR that gave a 90% success rate in each of the countries (at least I think that's what the "10th percentile" column is showing).

[arebelspy]

I'm happy to walk back the "made it popular" comment.  But there's no denying he's very influential on the topic.

Pfau is probably the most widely quoted guy regarding SWR research, mostly because he continuously updates the original research with new data.  His opinion is definitely more pessimistic lately.  Here's a catalog of his research.  https://ideas.repec.org/e/ppf6.html

I make no judgments on his methodology.  But he's put out some fine research worth noting and is willing to change his views with new information, and with a PhD from Princeton he's no internet hack.

I've been reading Pfau's blog for years; you don't have to convince me of his credentials.  I'm not sure what part of what I said translated to "who is this guy?"  :)

[arebelspy]

No, it doesn't assume it will be similar to the past--it assumes it will be no worse than the past. And those other countries worst case scenarios are worse than ours have been, yes, but I'm curious what their average scenarios were. That would be a lot more interesting to me, but I'm not aware of that data existing.

Pfau's study from 2010 on SWR's in international markets has some limited data on "average" scenarios.  In particular, Table 3 gives a bunch of data beyond the safemax WR's in various foreign markets, including the failure rates for a 5% WR (which was closer to being an "averagely successful" WR in some of the countries) and it also gives the WR that gave a 90% success rate in each of the countries (at least I think that's what the "10th percentile" column is showing).

You're right, this sort of has that for the ones that have around a 50% success rate of 5% after 30 years:
https://1.bp.blogspot.com/_7pcnNDkYOFE/TJLpRdCn9uI/AAAAAAAAAOI/e4WRVdW-D4U/s1600/Capture2.JPG

So that indicates to me that maybe a 4.5-5% success rate worked about half the time, even in other countries.  So 4% and flexibility should be fine, even assuming the US is worse than it's worst period, and is more like some of the other countries worse (but not worst) times.

[Tyler]

I'm not sure what part of what I said translated to "who is this guy?"  :)

Ha.  Nothing at all!  You're clearly well-versed in his background.  The longer explanation was more for the larger group who may not necessarily recognize the name.

[arebelspy]

I'm not sure what part of what I said translated to "who is this guy?"  :)

Ha.  Nothing at all!  You're clearly well-versed in his background.  The longer explanation was more for the larger group who may not necessarily recognize the name.

Gotcha.  :)

[TomTX]

I'm curious is the is reasonable, empirically researched data/studies that make a sincere or valid attempt at invalidating a 4% SWR.  In really dumb terms: the Sith to the Trinity Study's Jedi?

Sure.  By none other than Wade Pfau, the guy who made the 4% rule popular.

http://tinyurl.com/nrywjny

Pfau's recent work doubting the 4% rule have been based on some way or another handicapping expected returns.  I've seen two basic methods:

1) Sneak in a 1% management cost.  Pretty damn obvious. I and others called him on this one since most of us are with Vanguard and low cost.  So.... the next method:

2) Take historical returns. Cut them down 25-50%, then run the Monte Carlo. More subtle. Justified on current low bond yield. This is basically assuming we have times which will be considerably worse than the Great Depression, or 66-82 stagflation for investors. It's fine to play "What if?" - but this isn't really predictive.

[forummm]

From Brooklynguy:

ignoring the fact that it's arguably impossible to be "safer" than, say, 85-90%"

Let's be clear that isn't a fact, it's someone opinion (probably William Bernstein, a man who I respect).  Yes you can argue there is 10% or so chance you die before you hit 30 years,and 10% chance the economy changes and that works out to be 80 odd percent. But minimizing your chances of running out of money for that 80% of the time things go as expected isn't being overly conservative.

I agree with this and have noted before the irony in how Bernstein's 80% concept of false precision has itself become an example of false precision.

If this thread's purpose is to serve as a compilation of arguments about the 4% Rule and SWRs, we might as well add a link to the full thread where I made that observation, which had some good discussion on these matters in general:

http://forum.mrmoneymustache.com/welcome-to-the-forum/cfiresim-success-rate/

EDIT: Oops, just realized this is not the "Stop worrying about the 4% Rule thread."  But, for posterity's sake, clifp, maybe you want to respond to walt's identical post in that thread with the same response, and I'll do likewise to yours? :)

[forummm]

I'm curious is the is reasonable, empirically researched data/studies that make a sincere or valid attempt at invalidating a 4% SWR.  In really dumb terms: the Sith to the Trinity Study's Jedi?

Sure.  By none other than Wade Pfau, the guy who made the 4% rule popular.

http://tinyurl.com/nrywjny

Pfau's recent work doubting the 4% rule have been based on some way or another handicapping expected returns.  I've seen two basic methods:

1) Sneak in a 1% management cost.  Pretty damn obvious. I and others called him on this one since most of us are with Vanguard and low cost.  So.... the next method:

2) Take historical returns. Cut them down 25-50%, then run the Monte Carlo. More subtle. Justified on current low bond yield. This is basically assuming we have times which will be considerably worse than the Great Depression, or 66-82 stagflation for investors. It's fine to play "What if?" - but this isn't really predictive.

I too have found these methods by Pfau to be suspect. You could argue that he's aiming these studies at financial advisors, so the 1% is actually going to the advisor's pockets. But the Monte Carlo analysis has never been reflective of what has actually happened before. Sequence of returns get all crazy with that. In the real world, things don't behave that way.

[TomTX]

I'm curious is the is reasonable, empirically researched data/studies that make a sincere or valid attempt at invalidating a 4% SWR.  In really dumb terms: the Sith to the Trinity Study's Jedi?

Sure.  By none other than Wade Pfau, the guy who made the 4% rule popular.

http://tinyurl.com/nrywjny

Pfau's recent work doubting the 4% rule have been based on some way or another handicapping expected returns.  I've seen two basic methods:

1) Sneak in a 1% management cost.  Pretty damn obvious. I and others called him on this one since most of us are with Vanguard and low cost.  So.... the next method:

2) Take historical returns. Cut them down 25-50%, then run the Monte Carlo. More subtle. Justified on current low bond yield. This is basically assuming we have times which will be considerably worse than the Great Depression, or 66-82 stagflation for investors. It's fine to play "What if?" - but this isn't really predictive.

I too have found these methods by Pfau to be suspect. You could argue that he's aiming these studies at financial advisors, so the 1% is actually going to the advisor's pockets. But the Monte Carlo analysis has never been reflective of what has actually happened before. Sequence of returns get all crazy with that. In the real world, things don't behave that way.

They're certainly getting him headlines in the financial press (End of the 4% Rule!!!!!!) who use the titles as clickbait -  and this also feeds the frenzy over at Bogleheads with their "race to the lowest SWR."   Ignore Social Security. Ignore pensions. You cannot ever earn or recieve any other money (such as an inheritance. Or hobby income. Or churning credit cards for travel points. Or just Craigslisting your junk) You cannot vary your spending due to market conditions. Start with 3% as your SWR and you get a bunch of people arguing that it's too risky. You need 2.5%. No, you need 2%. No, I'm targeting 1.7%!

4% actually still seems pretty damn safe. Most of the risk is starting your retirement with a few bad years ("sequence of returns risk") - which can be planned for or solved pretty readily by a variety of methods:

1) I personally plan to have some "fun money" in the budget - primarily money for extended traveling.  The traveling can be postponed for a year or two in a market crash, or just made very cheap (ie National Park Camping Tour instead of European River Cruise)

2) In the first few years of retirement, you should be pretty darn employable or otherwise make some money. If the market crashes, work some for a year! (Yes, it's harder to get a job in a recession. Noted.)

3) I've got a lot of stuff around that I want to sell off. Don't have the time at the moment to do more than stay steady with selling off infant junk while acquiring toddler junk.

4) I'm gonna plan on Social Security. I plan to push it out to Age 70 (at the moment) - but if the market crashes hard, I can pull that in by 8 years. Frankly, being married there are a lot of middle ground options. Too complex for this post.

5) Personally, I have a pension.

6) I can have a HELOC in place shortly before I retire that I can draw on if need arises.

7) I am likely to inherit mid 6-figures within 15 years of retirement. (Perhaps morbid, but I am realistic here. The 15 years presumes one parent lives to 5 years older than any ancestor)

[Jeremy]

I'll give some props to Mr Pfau for highlighting one important (albeit somewhat obvious) point

The chart for Maximum Sustainable Withdrawal Rate shows a wide range.  4% was worst case.  But if you were lucky with your timing, it could be 12%



When we retired, were we on the 4% path or the 12% path?  No idea.  But we can get a pretty strong indicator early on

From Pfau, bullet point 3c, http://retirementresearcher.com/trinity-study-updates/
"Retirement success is more dependent on what happens early in retirement than late in retirement. In fact, the wealth remaining 10 years after retirement combined with the cumulative inflation during those 10 years can explain 80 percent of the variation in a retiree’s maximum sustainable withdrawal rate after 30 years."

If the sequence of returns in your first 10 years is favorable, and inflation hasn't gone crazy, then you are sitting pretty.  If returns aren't favorable, then you need to keep the portfolio large enough to benefit from the next bull market. 

How many people would retire with a planned 4% withdrawal rate, watch the market lay waste to their portfolio or their country plunge into a massive inflationary recession, and say, "No worries, the Trinity Study said 4% was perfectly safe, keep the champagne flowing!"  Probably not many

Small amounts of flexibility in spending (or earning) in the early years is probably worth a percentage point on withdrawal rate (+/-)

[a1smith]

Pfau's recent work doubting the 4% rule have been based on some way or another handicapping expected returns.  I've seen two basic methods:

1) Sneak in a 1% management cost.  Pretty damn obvious. I and others called him on this one since most of us are with Vanguard and low cost.  So.... the next method:

2) Take historical returns. Cut them down 25-50%, then run the Monte Carlo. More subtle. Justified on current low bond yield. This is basically assuming we have times which will be considerably worse than the Great Depression, or 66-82 stagflation for investors. It's fine to play "What if?" - but this isn't really predictive.

For some people the 1% fee is going to be valid (or low).  We just set up my wife's 401k for her part time job a few months ago and I just saw a 1% fee which appears to be taken out by the 401k administrator.  So, that is in addition to the ER's of the funds we have chosen.  The administrator just changed so hopefully those admin fees will be reduced and/or eliminated.  There's not much we can do about this fee; luckily my 401k has no admin fees and very low ER funds.

[forummm]

Pfau's recent work doubting the 4% rule have been based on some way or another handicapping expected returns.  I've seen two basic methods:

1) Sneak in a 1% management cost.  Pretty damn obvious. I and others called him on this one since most of us are with Vanguard and low cost.  So.... the next method:

2) Take historical returns. Cut them down 25-50%, then run the Monte Carlo. More subtle. Justified on current low bond yield. This is basically assuming we have times which will be considerably worse than the Great Depression, or 66-82 stagflation for investors. It's fine to play "What if?" - but this isn't really predictive.

For some people the 1% fee is going to be valid (or low).  We just set up my wife's 401k for her part time job a few months ago and I just saw a 1% fee which appears to be taken out by the 401k administrator.  So, that is in addition to the ER's of the funds we have chosen.  The administrator just changed so hopefully those admin fees will be reduced and/or eliminated.  There's not much we can do about this fee; luckily my 401k has no admin fees and very low ER funds.

Your 401k fee is not relevant for the SWR calculation. It does reduce your NW as long as you keep the money there. But once you've RE, you obviously won't be working anymore, so there's no reason to keep the money in that high-fee 401k. Roll it over to a Vanguard IRA at 0.05% and increase your SWR back to 4%.

[brooklynguy]

I too have found these methods by Pfau to be suspect. You could argue that he's aiming these studies at financial advisors, so the 1% is actually going to the advisor's pockets.

A more charitable view is that Pfau (and financial planners in general), like the 4% rule itself, are simply overcautious by nature -- they err on the side of caution because the consequences of being proven wrong are worse if their predictions are too aggressive than if they are too conservative.

[a1smith]

I too have found these methods by Pfau to be suspect. You could argue that he's aiming these studies at financial advisors, so the 1% is actually going to the advisor's pockets.

A more charitable view is that Pfau (and financial planners in general), like the 4% rule itself, are simply overcautious by nature -- they err on the side of caution because the consequences of being proven wrong are worse if their predictions are too aggressive than if they are too conservative.

I agree.  Most people absolutely, positively don't want to run out of money but don't mind having more left over to pass on to heirs, charities, etc.  Hence, the skewing towards conservatism.

[forummm]

I too have found these methods by Pfau to be suspect. You could argue that he's aiming these studies at financial advisors, so the 1% is actually going to the advisor's pockets.

A more charitable view is that Pfau (and financial planners in general), like the 4% rule itself, are simply overcautious by nature -- they err on the side of caution because the consequences of being proven wrong are worse if their predictions are too aggressive than if they are too conservative.

I agree.  Most people absolutely, positively don't want to run out of money but don't mind having more left over to pass on to heirs, charities, etc.  Hence, the skewing towards conservatism.

Funny, I actually I thought I was being charitable. Providing a reasonable justification why 1% fees could be baked into his analysis, when no DIY investor should have anything close to that. I assume that Pfau's target audience is financial planners, and they do tend to charge 1%--rightly or wrongly. For those advisor's clients, the SWR would be appropriately substantially lowered. For people like most of us here, we can happily increase our SWR by 25% since we aren't burning that money on fees.

I agree there's a skew towards conservatism in general. But a more informed approach would be to say the SWR is X% minus whatever fees your portfolio management has. That way if the advisor is actually charging 1% plus putting them in funds also charging 1%, the correct amount to let the clients live off of is only 2%. In this example, Pfau is actually not being conservative enough.

[Nords]

This reminds me of the saying "Everybody has a plan until they get punched in the face."

I think that people talk a pretty brave game about reducing their SWR below 4%... until they realize how long they'd actually have to work, or how much they'd actually have to save, or how much less they'd actually have to spend to live within their excessively-conservative artificial constraints.

The defense against failure is not reducing the SWR to 0.0%, and it's not determining the one true SWR out to six significant figures.  The answer is not abusing cFIRESim or FIRECalc until their success rate is 100.0000%.  The defense against failure is annuitizing a part of your portfolio so that you have a bare-bones income for longevity insurance or a zombie apocalypse.  Maybe that annuity is Social Security, maybe it's an SPIA, maybe it's a TSP inflation-adjusted annuity at age 70, maybe it's a military pension, or "all of the above".  But annuities are how you handle the questions about failure.  Hope you find a good insurance company to sell it to you!

Once you meet the boundary condition "assets = 25x annual expenses" then you're ready to figure out annuities and other asset-allocation questions.  You're ready to assess variable-spending schemes or throttling back to part-time work or side-hustle income.  You should not immediately revert to "assets = 33x annual expenses" or kick it up a notch to "assets = 40x annual expenses".  Unless, of course, you're willing to work that long for that amount of assets.

Funny, I actually I thought I was being charitable. Providing a reasonable justification why 1% fees could be baked into his analysis, when no DIY investor should have anything close to that. I assume that Pfau's target audience is financial planners, and they do tend to charge 1%--rightly or wrongly. For those advisor's clients, the SWR would be appropriately substantially lowered. For people like most of us here, we can happily increase our SWR by 25% since we aren't burning that money on fees.

I agree there's a skew towards conservatism in general. But a more informed approach would be to say the SWR is X% minus whatever fees your portfolio management has. That way if the advisor is actually charging 1% plus putting them in funds also charging 1%, the correct amount to let the clients live off of is only 2%. In this example, Pfau is actually not being conservative enough.

Wade and I had a Twitter exchange on that a couple weeks ago:
‏@TheMilitaryGuid  Jun 9
The 4% rule is "good enough", but study of variable spending:  http://buff.ly/1Ggvc67  H/T @WadePfau  Investors can beat an expense ratio of 0.5%!

‏@WadePfau
@TheMilitaryGuid I know, I know. But I'm trying to be representative for the general public, and that fee is still well below average.

[a1smith]

The defense against failure is not reducing the SWR to 0.0%, and it's not determining the one true SWR out to six significant figures.  The answer is not abusing cFIRESim or FIRECalc until their success rate is 100.0000%.  The defense against failure is annuitizing a part of your portfolio so that you have a bare-bones income for longevity insurance or a zombie apocalypse.  Maybe that annuity is Social Security, maybe it's an SPIA, maybe it's a TSP inflation-adjusted annuity at age 70, maybe it's a military pension, or "all of the above".  But annuities are how you handle the questions about failure.  Hope you find a good insurance company to sell it to you!

Once you meet the boundary condition "assets = 25x annual expenses" then you're ready to figure out annuities and other asset-allocation questions.  You're ready to assess variable-spending schemes or throttling back to part-time work or side-hustle income.  You should not immediately revert to "assets = 33x annual expenses" or kick it up a notch to "assets = 40x annual expenses".  Unless, of course, you're willing to work that long for that amount of assets.

Nice enumeration of options.

I had a similar, but less detailed, comment about people with long retirements (start at 30, 65 year retirement)

So, the safety factors MMM mentions in article listed above are not safety factors; they are an absolute requirement.

My main motivation in posting the cFIREsim data above was to show that the safety factors MMM mentioned are required, especially for long retirements, and not necessarily to reduce SWR.

For my own plan, I have pension and SS as annuitized income and potentially some inheritance funds.  Unfortunately, my pension won't have COLA like yours does.  I've used Fidelity's Retirement Income Planner to help set up my plan and it is variable SWR.  Some years I have >>4% SWR and some years I have <4% SWR due to timing of when pension, SS start.  This plan lets me retire sooner than having <=4% SWR at all times.

PS - It seems that we arrived at Pearl Harbor around the same time, I got there around the end of '81.  I was on USS NYC (SSN-696) till 12/83.  We're fellow nukes!  :-)  I really enjoyed the scuba diving when I was out there; unfortunately, most of my diving was on the sub.

EDIT - I meant to say I arrived at Pearl Harbor around the same time you started in the Navy, but you got the idea.

PPS - I am building into my plan a 75% reduction in SS benefits starting in 2033 as mentioned in Status Of The Social Security And Medicare Programs - A SUMMARY OF THE 2014 ANNUAL REPORTS

[TomTX]

Wade and I had a Twitter exchange on that a couple weeks ago:
‏@TheMilitaryGuid  Jun 9
The 4% rule is "good enough", but study of variable spending:  http://buff.ly/1Ggvc67  H/T @WadePfau  Investors can beat an expense ratio of 0.5%!

‏@WadePfau
@TheMilitaryGuid I know, I know. But I'm trying to be representative for the general public, and that fee is still well below average.

There is a MUCH better way.

"Fund expenses, management fees, advisor fees, etc count as part of your withdrawal"

Might get more of the general public to wake up that their advisors are taking perhaps 50% of their spendable money every year.

Yep. Skimming off half for usually doing worse than the indexes.

[Nords]

I've used Fidelity's Retirement Income Planner to help set up my plan and it is variable SWR.  Some years I have >>4% SWR and some years I have <4% SWR due to timing of when pension, SS start.  This plan lets me retire sooner than having <=4% SWR at all times.

I think that someday the research (and computer simulations) will catch up with reality to verify that the 4% SWR with a variable withdrawal plan has been the right answer all along.

PPS - I am building into my plan a 75% reduction in SS benefits starting in 2033 as mentioned in Status Of The Social Security And Medicare Programs - A SUMMARY OF THE 2014 ANNUAL REPORTS

If the system performs better than the GAAP estimates, or if the payroll contribution cap is removed, then everyone will breathe a lot easier.

PS - It seems that we arrived at Pearl Harbor around the same time, I got there around the end of '81.  I was on USS NYC (SSN-696) till 12/83.  We're fellow nukes!  :-)  I really enjoyed the scuba diving when I was out there; unfortunately, most of my diving was on the sub.

Wow-- not only that but we're shipmates.  I was NYC's Weps from Dec '89 to Jul '92.  (I spent '82 through '86 in the Atlantic, and '87-'89 at Monterey.)  I got the Weps job because the incumbent had resigned his billet for psychiatric reasons, and the crew (let alone the CO) was a little skittish about me coming onboard as his replacement.  NYC's Engineering dept graciously allowed me to stand my proficiency watches (and give the young O-3s their share of OOD time) but of course I was still occasionally regarded as a brain-damaged nuke stepchild-- just as if the nukes were treated if they'd wandered into sonar or the torpedo room.  It was definitely not an easy tour but in retrospect I did what I needed to do, helped a lot of people out, and got what I needed to get.

I did my diving in ballast tanks and occasionally sanitary tanks.  In retrospect I wish I'd started my career in Pearl Harbor and on SSNs, but coming to that community after Holy Loch and SSBNs made me appreciate Hawaii Navy ever so much more.

[ender]

The defense against failure is not reducing the SWR to 0.0%, and it's not determining the one true SWR out to six significant figures.  The answer is not abusing cFIRESim or FIRECalc until their success rate is 100.0000%.  The defense against failure is annuitizing a part of your portfolio so that you have a bare-bones income for longevity insurance or a zombie apocalypse.  Maybe that annuity is Social Security, maybe it's an SPIA, maybe it's a TSP inflation-adjusted annuity at age 70, maybe it's a military pension, or "all of the above".  But annuities are how you handle the questions about failure.  Hope you find a good insurance company to sell it to you!

This is one reason I would like to have some rental property income by the time we RE.

It's not as "guaranteed" as a pension/annuity but it would be nice to have a portion of FIRE income coming from rental properties instead of portfolio withdrawals.

I will have a small (slowly growing...) pension as well as, so we'll have some combination of SS, private pension, tax advantaged, and hopefully real estate. It'd be nice to have real estate investments covering 50% of our expenses - though I'm young enough I have no idea what life will bring for our FIRE timeframe.

[forummm]

The defense against failure is not reducing the SWR to 0.0%, and it's not determining the one true SWR out to six significant figures.  The answer is not abusing cFIRESim or FIRECalc until their success rate is 100.0000%.  The defense against failure is annuitizing a part of your portfolio so that you have a bare-bones income for longevity insurance or a zombie apocalypse.  Maybe that annuity is Social Security, maybe it's an SPIA, maybe it's a TSP inflation-adjusted annuity at age 70, maybe it's a military pension, or "all of the above".  But annuities are how you handle the questions about failure.  Hope you find a good insurance company to sell it to you!

This is one reason I would like to have some rental property income by the time we RE.

It's not as "guaranteed" as a pension/annuity but it would be nice to have a portion of FIRE income coming from rental properties instead of portfolio withdrawals.

I will have a small (slowly growing...) pension as well as, so we'll have some combination of SS, private pension, tax advantaged, and hopefully real estate. It'd be nice to have real estate investments covering 50% of our expenses - though I'm young enough I have no idea what life will bring for our FIRE timeframe.

There's no difference between selling appreciated stock and spending the dividends. Real estate is imperfectly correlated to stocks, so it's one way to decrease volatility. But so are bonds. REITs are another way to own real estate with a diversified portfolio, no liability, be highly liquid, etc.