We made it through October, without much volatility this time – what a change compared to last year! We even got to a new all-time high in the S&P 500 in the last few days. When you reach new records, the pessimists come out of the woodworks and declare that “this is the top” and the next bear market must be right around the corner! It’s like clockwork! And if you go to the popular forums and Facebook Groups in the FIRE community, you’ll see people poking fun at the perma-pessimists. Quite appropriately, I think!
But why are people still a bit nervous about corrections and bear markets and market crashes? Being retired now, I have to admit I feel at least a little bit uneasy right now. Why’s that? If I wanted to quantify how afraid I am of something I’d do so as follows: Fear depends on both the probability and the magnitude of something scary happening: FEAR = The probability of something scary TIMES the magnitude of something scary
In my recent post My thoughts on the “Upcoming Recession” I wrote about the probability part. I personally don’t think that the economy is at the brink of a major slowdown (yet) and with the economic growth trend, still intact the stock market will likely chug along. This all looks like a mid-cycle, temporary soft spot.
What makes me nervous about the bear market prospect, though, is the magnitude part; the fact that IF a bear market were to occur (however unlikely that may be) we’d most definitely go through some anxiety for a while. That’s true for all retirees and even folks close to retirement. Probably not so much for everybody just starting out in their accumulation phase, see the post “How can a drop in the stock market possibly be good for investors?” from earlier this year.
Quite intriguingly, though, if you look around in the FIRE community I get the sense that people minimize how scary a bear market will be if it were to start today. And the thought process is:
- The market will always recover (see the chart above)
- Most bear markets last only about one to two years
It sounds like the bear has really lost its teeth! So, why am I not convinced? There are multiple problems with that line of thinking. That 1-2 years estimate wildly underestimates how long it takes to recover from a bear market. If you do the math right a bear market will appear a lot scarier than it’s commonly portrayed. Let’s see why…
Problem 1: Ignoring the time it takes the market to recover
The average bear market may last only about 1-2 years. That doesn’t sound so bad. But what people routinely ignore is that the length of the bear market is defined as the time from peak to trough only. How long will it take to get back to the old peak? Well, let’s look at an example. Here’s the cumulative return chart for the S&P 500 index (Total Return, so dividends are reinvested) after the year 2000 market peak. I do this whole analysis based purely on monthly data, so with the intra-month highs and lows, you might get slightly different results but qualitatively still very similar! The bear market took 25 months. But it takes another 4 years to just get back to the old peak! That means your portfolio was underwater for more than 6 years.
OK, that was one recession and one bear market. How about others? In the table below, I gathered data for 10 major market drawdowns (20%+) since the Great Depression: The market peaks occurred in 1929, 1937, 1946, 1961, 1968, 1972, 1980, 1987, 2000 and 2007. (side note: I’d consider February 1937 a new peak because we came really, really close to a new peak if you account for dividends and the CPI adjustments, which were negative due to deflation in the 1930s. The price index in nominal terms didn’t even come close to recovering until much later!!!)
I report this both for the S&P 500 price index but also for the more meaningful Total Return Index, i.e., with dividends reinvested. Notice how the average bear market took around 16 months, but from peak to trough to a new all-time high took 56 months on average. About 3.5 times longer than the initial bear market. It seems the stock market is about 2.5-times slower on the way back up than on the way down!!!
So, just right off the bat, that mantra “nothing to worry about, the bear market will be over quickly” seems like a bit of false advertising. You’ll be underwater for an average of almost 5 years, instead of one to two years! And remember, from a Sequence of Return Risk perspective it’s the time you spend below the peak that will create problems with your retirement strategy, not just the length of the equity drawdown!
But it gets even worse, which brings us to the next point:
Problem 2: Ignoring Inflation (again!!!)
Recall my post from a few months ago: “How To “Lie” With Personal Finance” where Lie #2 deals with the way we sometimes delude ourselves if we ignore inflation. If your expenses increase during that time due to inflation, you want your portfolio to not just catch up with the old all-time high, but with that all-time high plus inflation. How long would that have taken after the year 2000 stock market peak? That’s in the chart below. Shockingly, the 2002-2007 bull market was only enough to reach a new high in 2007 in nominal terms. But that level still had a negative real rate of return; the blue line never gets back to the yellow line! Then the 2007-2009 recession and bear market hit and took your portfolio to the woodshed again. You would have returned to your August 2000 portfolio level only in April 2013(!) when adjusting for inflation! Almost 13 years later.
How about the other bear markets? In the table below, I report the results for the Total Return Index. So, 82 months on average, just under seven years. And that’s the average. Around the really scary bear markets (1929/1937, the 1970s and 2000) we’d spend anywhere between 12 to 15.5 years underwater.
So, picture yourself as a buy-and-hold investor having to wait 5 years (or 10+ years in the bad bear markets) for your portfolio to recover. It wouldn’t necessarily ruin you financially, but missing out that many years of market gains is a major opportunity cost. Nothing to take too lightly!
And you’re even worse off if you’re a retiree, due to Sequence of Return Risk! And talking about retirees, this spells trouble for all the folks who claim that the simple solution to Sequence Risk is to just keep a “cash bucket,” i.e., a certain number of years worth of expenses in non-risky assets (money market, CDs, government bonds). Well, unfortunately, 16 months’ worth of expenses – the length of the bear market – will not suffice. 56 months of expenses (back to the new all-time-high) will not suffice. You’ll need about 82 months of expenses to cover just your average bear market, significantly longer in the really bad bear markets!
But even that 82-month cash bucket is not enough, which brings me to the next point…
Problem 3: Ignoring that a 0% real return still really, really sucks!
Well, OK, the recovery takes a little longer than the 1-2 years normally floating around. But 82 months is still not that long, right? We just have to keep about 7 years worth of expenses in the cash+bond buckets and then we’ll be OK. With a 4% withdrawal rate that makes 28%. So, why not have 72% in stocks and 28% in a money market fund? Then we should be completely protected from Sequence Risk if an average bear market hits, right?
Wrong! The bucket proponents routinely miss one crucial detail! Let’s look at the following example: Imagine you start with a $1,000,000 portfolio. You put 72% or $720,000 into equities and 28% or $280,000 into your cash bucket yielding about enough to keep up with inflation (not a bad assumption right now). You withdraw $40,000 a year adjusted for inflation from the safe asset bucket. You never had to touch your equity portfolio while it was underwater. So, why haven’t we succeeded in hedging against Sequence Risk? Well, if the equity portfolio only just returns to its old all-time high in CPI-adjusted terms after 7 years, we now have an equity portfolio worth $720,000 (in year 0 dollars) but we’re withdrawing $40,000 p.a. (in year 0 dollars).
Do you see a problem with that? We now have a 100% equity portfolio and a 40/720=5.56% withdrawal rate! Simply returning to a new all-time high in your equity portfolio is not enough, not even if adjusting for inflation! You also have to replenish your diminished cash bucket before the next big bear market as well! So, to make it back to a $1m portfolio, your equity portion should have grown by roughly (1000/720)^(1/7)-1=4.8% over and on top of inflation to make it back to the initial portfolio level. In other words, just to keep up with the withdrawals from the cash bucket, my equity portfolio has to work much harder than 0% for 7 years. You need about a 5% return after CPI inflation!
So, let’s see how long it would take for the equity portfolio to not just recover in real, CPI-adjusted terms but even add roughly 5% per year. So, we take the same chart as before and add another line “CPI+5% p.a.” in green, see below. Bummer! After the 2000 market peak the S&P 500 never even got back to the old trend path! 229 months and counting! Ouch!
Well, the good news is that, during all of the other bear markets, the S&P 500 indeed managed to catch up to the CPI+5% growth path. But in some cases, it took forever. I mean F-O-R-E-V-E-R! 26 years in the Great Depression. 27 years in the 1960s! And these are, no coincidence, exactly the recessions and bear markets where the 4% Rule failed and even the cash bucket strategy would have had trouble saving the 4% Rule.
Talking about bucket strategies, the discussion wouldn’t be complete without mentioning my good blogging friend Fritz Gilbert over at The Retirement Manifesto who is a big fan of a bucket strategy. But, to my knowledge, Fritz doesn’t claim that the bucket strategy will ever solve Sequence Risk. It may alleviate it a bit but never eliminate Sequence Risk. That’s consistent with my findings both in my Safe Withdrawal Rate Series. See Part 19, Part 20 (glidepaths, i.e., liquidating bonds first during the bear market, so very similar to a bucket strategy) and Part 25 (some explicit cash bucket simulations). So, just to be sure, when I say, some people are wrong about bucket strategies, I wasn’t talking about Fritz! 🙂
Summary
Again, I’m not saying a bear market is around the corner. It’s a low-probability event. But IF a bear market were to occur, it’s a lot scarier than people want to make it. True, the market has always recovered! It will also recover following the next bear market. The bad news: it’s not really a matter of if, but when! If the recovery takes too long it spells trouble for investors and especially retirees! For retirees with an aggressive withdrawal rate, it might make the difference between making it and running out of money. If you have a more conservative withdrawal rate – yours truly included – it might make the difference between having tons of money left over and just scraping by.
The average bear market may last for only 16 months, but the recovery will last a lot longer. The anatomy of the average bear market looks like the one below. For the severe bear market events (1929, 1937, 1968, 1972, 2000), expect to double the time it takes to recover.
If you’re not scared yet then I don’t know what it would take. But, of course, there’s always Halloween right around a corner…
Are you scared about a bear market? Please leave your comments and suggestions below!
Title picture credit: Pixabay.com
Thank you Big ERN! This helps those in the accumulation mode keep their nose down and dollar cost average the whole way through your much more realistic picture of a downturn than what we typically tell ourselves. And those getting close to needing withdrawals from your accounts, De-Risk already!
Strategies to de-risk?
Glidepath, Bond Tent. See SWR Series Parts 19 and 20.
Hold GOLD 🙂
Not a bad idea! If you keep it a low enough % you might find that a useful hedge against Sequence Risk! 🙂
Would like to hear your strategies on de-risking.
SWR Series Parts 19-20
Ha, yes. Lots of gloom and doom and “why bother” here on a site devoted to retiring early through investing. I’m having a hard time determining if ERN is pro or anti early retirement
If we lived in a black/white or 0/1 binary world, I’d be pretty hard to pin down.
I’m pro retirement. But be prepared for a drawdown.
I’m way hip to that. I get accused of being too cautious all the time. Having 27 1/2 yrs vs 25 yrs of assets and keeping some cash sounds pretty reasonable to me but is sacrilege to most others. (The dreaded “Cash Drag!”) Most of these “invest your way to early retirement” forums/blogs are way too cheery in my opinion. “Draw down? So what? It always comes back.” At least you’re getting it –ALL– on the table.
Haha, thanks John. Well said! Retirement is more complicated than accumulation. Much worse Sequence Risk!
Yeah, in the accumulation phase it may not look quite as scary. But there’s a rationale to go to somewhere around 60-80% stocks by the time you retire.
Very scary indeed!!
So, other than being aware now of just how bad it could be, and soon to be retiring, what should we do to prepare? 50% of our retirement income comes from a guaranteed indexed company pension, the rest is from a self-directed 70/30 portfolio – we have 3 years cash + 3 years in bonds. And I thought that was more than enough! The company pension is enough to cover 75% of our FIRE needs.
Wow! Consider yourself lucky. You should not be scared of a bear market (unless the bear is bad enough and takes down your pension). 🙂
Were ready, going to buy everything when it does
Haha! I hope you didn’t keep all that cash since the S&P was below 2000 though! 🙂
I’ve been beating this same drum forever. Nobody listens. You’re absolutely right of course — Bear markets can last a very long time until a full recovery happens…in some markets (like Japan) it’s been 20+ years without a full recovery. There might never be one!
Great point! Yeah, Japan is a whole other level of scary. Three lost decades in the stock market. I’m hopeful though, that we’re not falling into trap because the U.S. market was never as overvalued as Japan in the late 80s, early 90s.
How graphs look for 3.5% and 3%?
At 3% and 3.5% you would have caught up to the year 2000 high already.
3.5%: in 2018! Still a long time with crummy returns
Can you guys clarify if you mean a floating 3% or an inflation-adjusted value corresponding to an initial 3% WR?
I’d also really like to know how this would have worked out with a CAPE10-based SWR (a=1.0%, b=0.5) given the previous installments of the series. The initial WR would have been quite low in several of the examples (and 🙁 now).
I think the 3% referred to “how quickly does the S&P catch up to CPI+x%” and the x% in the post was 5% and someone asked what about 3% or 3.5%. It didn’t mean a SWR, the way I understood it.
Yeah, I’m a big supporter of targeting a stock-valuation-based WR. See Part 18 in the SWR Series. A CAPE rule would have done really well in the 2000s.
Ok, thank you.
Fritz, great article but wouldn’t the picture look a lot better if you used a globally diversified portfolio instead of one concentrated in one market sector (US Large Cap)? No responsible planner would ever suggest concentrating all equity into one narrow portion of the market. Why do you do that? I assume it may be because the data is so readily available for the S&P 500, but in taking that shortcut aren’t you creating more fear than really exists?
Sorry BigERN, your post made me have Fritz on my mind.
Haha, gotcha. Calling a German Fritz? Hmmm, I’m glad we all have a great sense of humor here! 🙂
Global diversification will not be very useful in a bear market. All markets tanked in the Great Depression, the 70s, 2000s and 2008/9. Sometimes even worse than the U.S.
Yes, they fell, but not as much and they recovered much more quickly. Has there ever been a 10 year period when all equities were down? For the S&P 500, it’s about 15% of the time. By diversifying you can eliminate much of the risk to your fixed withdraw or bucket strategy.
Rule of thumb: in the recession and bear market every single market will drop.
Recoveries can be very different, see this post:
https://earlyretirementnow.com/2017/08/23/how-useful-is-international-diversification/
So, sometimes diversification accelerates your recovery (e.g. post 2001), somtimes it slows you down even more (post 2008/9). It’s a crapshoot!
What you allude to here is ye ol’ Recency bias… and most of us have trouble fathoming a Bear because market recently keeps bouncing back… until it doesn’t : ) Thank you for the healthy warning!
Yeah, we were spoiled the last 10+ years. And people tend to forget after that many years. Very true! 🙂
Excellent post, well thought and researched, as always. While I understand the principles here, was hoping for strategies if an extended bear market occurs or is it the standard FIRE principles of decrease unnecessary spending, geo arbitrage, side hustles, etc?
take a look at his glide path articles. de-risk by leaning towards bonds in the first ~10-yrs of retirement. another strategy a lot of folks follow is to diversify into physical real-estate (single family homes, multi-family, commercial) (actually owning/managing vs. REITs; or like Ern consider Syndicates)
Yeah, all great ideas!
I guess options are another hedging strategy (which ERN employs as well). They earn money no matter which direction the market takes.
True, as long as the market deflates in an orderly manner. Great point!
https://earlyretirementnow.com/2019/03/27/passive-income-through-option-writing-part-3/
Oh, I’m just pointing out the the numbers. What you do about it in your budget is up to you. i guess the new ChooseFI book by Chris Mamula is a good start.
Also, if you’ve already maxed out everything and done all the geo-arbitrage etc., you’d have to find even more savings potential. May not be that easy… 🙂
ERN, you’re once again proving why you’re one of the top bloggers in the world today. Thanks for the shout-out, and glad to hear I’m not entirely wrong with my bucket strategy. For the record, I do agree that the bucket approach doesn’t eliminate Sequence of Return Risk, it’s really just an asset allocation strategy that helps identify why the various allocations are represented in the portfolio.
I do wonder about the potential super-bear phenomenon of both bonds and stocks potentially being over-valued. My focus is on a highly diversified portfolio (REITS, Int’l Equities & Bonds, Small Value, Commodities, etc), with the hope that there will be something I can sell in a worse-case scenario to refill bucket 1. And, of course, we’d all likely reduce our withdrawals/spending if we faced a massive bear.
Thinking about it, the answer is clear. Go to 100% Cash right before the Bear, then move to 100% Equities at the bottom. Can you put a post together that tells me when to make those moves? Then you’d TRULY be the best finance blogger in the universe. Tag me when that one’s out, I can’t wait.
Great post, as usual. I would like to see a follow-up on your recommended contingency plans, I know you’ve written on it in the past but think a focused post on “So, What Do You Do About The Bear Scenario” would be a natural follow-up.
Great post as always Big ERN. As Fritz stated, you have quickly climbed the ranks as a premier finance blogger. I always smile when I see you’ve posted again because I know I’m going to get schooled.
Also, please tell me as well how to time the market LOL!
Finally, three main ways to counter this event: save more than your “number”, obtain multiple streams of income (think real estate or business) and / or go back to work. Keep in mind, going back to work may be harder than it sounds during a significant bear market…
Keep up the great work my friend!
Thanks, Frankie!
These are all great suggestions. And I completely second that caveat of yours: going back to work during a recession is not that easy!
Thanks Fritz! Whew, I’m glad we’re on the same page!
I’m concerned about the Super-Bear as well. Talked with Todd Tressider at FinCon19 about it. Of course, if you calibrate your withdraw strategy to historical data and you would have made it through the 1970s/80s, the mother of all super-bears because stocks AND bonds tanked, you should be safe for what’s ahead. At least I would think so.
Market timing? Yeah, I wish I could do that reliably. But then I’d probably keep it to myself and not share it! 🙂 But a Glidepath from 60/40 to 80/20 would probably be a start.
Good suggestion for a follow-up post. I will think about that!
Yeah, I want a post that tells me when to go 100% cash at the “real” market top too, and when to jump back into equities at the “real” market bottom.
I live in Las Vegas, so I have other money making opportunities besides the stock market. If you have any posts on when to bet on black versus red at the roulette wheel, that would be helpful too.
But seriously, thanks for the great post as always!
Haha, good one. Trying to predint the top is just like predicting the roulette wheel. 🙂
As someone who retired last year at 59, I rode out 1986, 2000, & 2007 taking advantage of dollar cost averaging, therefore, I am somewhat confident that my retirement portfolio will survive. However, a downturn will be unchartered waters for me and most who recently retired. I wonder how the not so recent retirees have fared when retiring in 1970s, 80s, and even 90s. This is when I use Prof ERN’s SWR Toolbox to assuage my concerns by running the gamut of retirement month/year possibilities with my portfolio. Thank you for crafting such a valuable tool. Its better than paying a copay to a shrink! In a world of instant news, just stick to an achievable plan. Ignore the noise. I remember driving in my Army Humvee in October 1986 knee deep in Mount St. Helens ash, in my hair and nose passages and finding a workable phone booth in the middle of Yakima Firing Center in WA east of where Prof ERN lives. I had no laptop, internet, smart phone etc for instant news. I called my then wife who told me that the market crashed. I did not panic after getting what was “late news”, I simply kept to my investment plans in tact and even bought more shares when I returned from deployment and rode it out.
Wow, that’s an amazing story! You’re right: decumulating is even harder than accumulating assets! Better not lose your nerves now! Glad you found my tools useful! They gave me the confidence to retire! 🙂
This is where I wonder if using a percent or so of capital to carry some long dated puts as insurance would help? I’ve done that in the past and invariably they end up profitable. Would like to see it back tested though, it slows me to carry a higher percentage of equities and actually provides a hedge. Like Fritz mentions, bonds and stocks could go down together.
I like the idea of doing long-dated long calls and investing the rest in some low-risk interest-bearing assets. Looks like an ATM call costs around 9% premium for an expiration 2Y out, so 4.5% p.a. Might be worth it!
Predicting the start of a bear market is near-impossible, but one generally knows it when markets are down 30-40%. That near-bottom is also generally the all-clear sign indicating a cohort when a 5-6% WR is sustainable.
Perhaps an AA and starting WR that would lead to such a WR in the event of a standard bear market should be the target? E.g. if bonds don’t change and stocks drop 40% in the modeled bear market a $1M/4% plan changes to a $750k/5.3% plan. And maybe that’s OK because risk is a lot lower at the bottom than it was at a not-bottom.
Or for a person a couple years from retirement in an environment with a high CAPE, deflationary pressures, and deteriorating demographics, might it make sense to hide in a bond-heavy AA for a couple of years “waiting for the sale” so to speak? Particularly considering that contributions will have a bigger and more certain effect on reaching one’s FIRE number than investing returns, why risk early derailment when one does not yet need to live off the funds?
I’m also considering a 90% bonds, 10% long calls (S&P LEAPS) strategy that would switch to a standard 90/10 if stocks fell X%. Today’s low VIX makes that a tempting move to take risk off the table at a bargain price.
Sure, when the market is down you can use a higher SWR because valuations look more attractive. That’s what a CAPE-based SWR would be.
But caution: your withdrawal amounts will still vary over time. You will still encounter prolonged drawdowns in withdrawal amounts.
I’m also intrigued by LEAPS. Premiums for ATM call options really far out (2 years) are pretty affordable. I’ll think about how to implement that into a SWR analysis!
Are you thinking calls because you expect yields to fall further and a melt-up rather than a melt-down? I’d be interested in your line of thinking if you feel like sharing
No, the Call strategy is just a cheap insurance againt the really bad drops. You don’t really need to take a view on rates to find this interesting. 🙂
The long LEAPS calls guard against the risk of being wrong and watching from the sidelines as the market takes off. This or a protective put strategy are ways to get unlimited upside but set a floor on losses.
Add selling weekly calls against that long LEAP and you can recover those 4.5 %.. Some pretty knows hedge funds do something similar. I really like this stratety. This way you are long slowly decaying asset and short fast decaying one. You can sell 50 x weekly calls so you only have to recover 0.08 % each week. Or because 2 days out options are more your cup of tea, you can sell those 2 days out. 150 of them to cover the yearly cost of long call..:)
Yeah, I like this calendar spread strategy, because the premium is so much higher for short-dated options! 🙂
If you want to live in fear, that’s up to you, Big ERN. Unless you bought all your equities the day before a market peak, why should it make any difference how long it takes for the market to return to that arbitrary point?
It makes a huge difference. If you retire with $1m now and withdraw $40k you’re in trouble if we have a bear market. Explain to me how it makes a difference if you bought that $1m today or you bought it for $100k many years ago.
If anything, if you bought into the market at a higher basis you can use tax loss harvesting to cushion the fall. So, in that sense, the damage would be a little bit less severe if you just bought your portfolio at the peak.
Agreed, if you retire with a $1MM portfolio today from which you *must* spend at least $40K/yr, and the market crashes tomorrow, you’re in trouble. Given all of the good advice in your SWR series, why would anyone reading your blog possibly do that? You’re certainly not doing that. I’m not. So, where are all these imaginary people who are retiring with $1MM and inflexibly expecting to be able to spend $40K/yr? If these people actually exist, I agree with you, Karsten, they should be afraid, but the rest of us? Not so much. By definition, most people will *not* retire the day before a market peak. It only matters how long it takes for the market to return to the inflation adjusted price of stocks on the day you retired, *not* an arbitrary market peak that happened years after retirement.
I completely disagree. It matter at any point when a bear market starts during your retirement years. Who cares if you’ve been retired 10 years ago or just today? From the moment a bear market starts in your retirement, you will take an average of 13 years to get to the same inflation-adjusted net worth while recouping the cash withdrawals you needed to sustain your retirement. ERN ist right, don’t take this lightly.
I talked about the flexibility myth in SWR parts 23-25. The EXACT SAME fallacy is floating around there: “you have to be flexible only for 1-2 years, for the duration of the bear market.”
But that’s a myth. So, you want to cut your withdrawals for 7-10 years? I have some examples where flexibility would require 20 years of reducing withdrawals by 30% (or alternatively getting a job).
I’m not that flexible. Are you?
Again, Karsten, I’m totally agreeing with you that it is risky, due to SORR, as well as other reasons, to retire with exactly $1MM and the expectation that you will be able to inflexibly withdraw $40K/yr, no matter what. I think that this point you have made repeatedly is a valuable contribution to the FIRE Community. Also, it’s not my point, at all, to claim that a bear market is not bad or is somehow wonderful for investors. Obviously, that’s not the case. All I’m saying is that for investors who are already following your and others’ advice to plan on spending less the 4%. A 3% – 3.5% WR seems much more prudent, especially now, given high CAPE values, and who have the flexibility to reduce or eliminate sales of stock, during downturns, there is ABSOLUTELY NO REASON TO WORRY or “BE AFRAID” of a bear market. They’re normal and to be expected.
Imagine a person retired in early 2016 with 10K shares of VTI that he bought for $100/share, i.e. $1MM. Today, those 10K shares of VTI are worth ~$1.55MM. Imagine today is the market peak and VTI drops by 40% to $93/share, obviously that would be really bad, but an investor who retired with $1MM of VTI would only need to wait until the share price of VTI recovered to its inflation adjusted price on the day he retired. You are completely richtig that the purchase price of the shares is unimportant. I’m arguing that the price of VTI at the market peak is also completely irrelevant. Who cares how many years it takes for the price of VTI to recover to $155? What matters is the price of VTI on the day you retired. For a retiree who purchased VTI just a little over 3 years ago, if the market crashed by 40% today, he would only need to wait until the price of VTI recovered to ~$107/share, i.e., the inflation adjusted price on the day he retired in 2016. Also, as @Chris B says immediately below in this thread, once the market crashes, CAPE will fall along with it, which will most likely make at least a 4% WR safe, once again.
Does that make any sense to you, Big ERN? BTW, I enjoy reading your blog. Thanks for all of the work that you do. Hope you’re enjoying your new life in Cali. Looking forward to reading more of your posts.
Shane, for what it is worth I agree 100% with you here. Been retired four years now and the value of our assets is considerably higher now than when I retired. We have a conservative asset allocation (40% equities 60% fixed income). So, a 40% drop in equities by itself only reduces our overall portfolio by 16%. This changes nothing in terms of our annual spending. Even a 50% drop in equities only translates to a 20% drop in our portfolio. Yes, this would bring our portfolio below its value on the date I retired, but not by enough to modify our withdrawal plan.
Speaking of which, we are following a modified version of Guyton & Klinger’s decision rules (modified to start with a much lower initial WD rate (3.25%) and customized guardrails based off that IWDR). Annual WD dollars set in initial year or retirement and increased annually for inflation. Current WDR below 3.0% due to growth in portfolio value.
In addition, a sizable % of our spend is variable (based on T&E). So we could fairly easily pare back spend in the frightening environment being talking about here and we would be fine. Worst come to worst, we could downsize from our current (larger) home for a more permanent step change in our spending level.
I think the issue raised in this thread applies more to people who retire shortly before a severe and prolonged bear market (as you have rightly pointed out) and who don’t have a lot of room to pare back spending (and/or began their retirement using a high(er) IWDR).
OK, gotcha. We’re on the same page, 100%, it seems. Whew! 🙂
Let’s be careful not to double-count our risk. Once the SORR has already occurred, I think history shows a second event is less likely for several years. A 3.5% WR on a 80% equities portfolio was riskier in 1999 than a 4% WR in 2002 – maybe even 4.5%.
To say nothing less than a short term return to a 4% WR is acceptable is to say the risk of back to back market crashes is the same as the risk of a single event, when in fact this scenario is much rarer.
I never doubted that. I only said that it if your $1m portfolio and $4k p.a. withdrawal gets into trouble it doesn’t matter if that $1 was a one-time investment or hs been accumulated over many decades with built-in capital gains.
Thanks Karsten, just in time for Halloween 😉
What do you think about these work-arounds?
– calculate your nest egg as a 200 week average (leads to a smoothed, lower nest egg -> takes longer to reach multiples of 25 of your spendings)
– retire not at 25x annual spending but at 33x spending
– after 7-10 years bear market, social security/pension comes more in sight (you need those 7-10 years less in your nest egg, not the same amount?)
– retire not in an up market but keep working till a bear market (S&P500 ~30% down from top)
– don’t worry, be happy 🙂
LG Joerg
200 week? You mean 200-day? Yeah, that would prevent retiring based on a short-term fluke.
33x might be a little too conservative
Oh, yes, Social Security will kick in. But I already factored that into the withdrawal analysis and raised my SWR from 3.25 to 3.5%.
Yeah, if you wait for the next bear market you might also a) wait really long and/or b) due to your decimated portfolio but constant expenses retire with a WR much higher than 4%.
I like the last one the best! 🙂
Thank you Karsten,
Indeed, I meant 200 week average.
Your “Halloween Trick” is based on calculations from an overvalued peak (Aug.2000, a bubble, as we know today).
The 200-week average of a deposit value in August 2000 (ie ~ 4 year average) would have been correspondingly lower (25-33% lower?), so that a 25x FIRE target would have fit from the lower level?
So, either first calculate yourself poor (eg 200 week average), then it will be enough? Or count on the current sum of 30x (instead of 25x).
To be relaxed, if after 10 years the initial depot height is no longer reached due to a bear, your lifetime is also ten years shorter (so you wouldn’t need the same CPI adjusted amount any more)?
LG Joerg, I do appreciate your blog very much!
I don’t think that we are in the same overvalued situation as in Aug.2000 due to the lower interest rates.
The opposite can come true as well (surging stock markets arround the world – may be less in US?)
But that is dumb speculation 😉
All good points!
You would also re-compute everything after 10 years of retirement. If your initial horizon was 30 and now it’s 20 that will make a all the difference. Not so much from 50->40 years, though, so this bear market risk is a lot more scary for early retirees.
Yeah, 2000 seems way worse than today. But note the 1960s which are roughly where we are today!
I hear you Ern, may I know what is the solution if cash bucket won’t work and it is not possible to time the market/predict the next cash?
I meant the next market “crash”.
What if we only start using the cash bucket after the market crash above 10% , will that help?
Tried many different variations of the cash bucket (e.g. in SWR Series Part 25: use cash bucket if portfolio is 20% underwater) and, no, that’s not going to work either. You just don’t have a large enough cash bucket:
https://earlyretirementnow.com/2018/05/23/the-ultimate-guide-to-safe-withdrawal-rates-part-25-more-flexibility-myths/
I have to admit that I’m a bit confused by your comment: In Part 25 you were pretty positive about the Cash Bucket but in your comment above you say it’s “not going to work either”. Is your comment only related to a 4% WR or targeted in general towards the Cash Bucket?
The cash bucket can somewhat alleviate the SoRR. But it’s no panacea against running out of money with a 4% Rule.
That makes sense. As I never planned with 4% but with a much lower percentage it is good to hear that I can use something like a cash bucket for better sleep 🙂
Likely a lower WR than the 4% for most extremely early retirees. And use a glidepath. There is no true solution to the Sequence Risk problem.
You are my go to numbers guy. Thanks!
Thanks! You made my day! 🙂
Shane, I think you are missing the point. If, like most in the FI community; you retire at exactly your number, this has huge consequences.
If you retire at “exactly your number,” exactly at the peak, then yeah. Otherwise, not so much.
You are exactly right. I can’t argue with that. I just think he was talking about the possibility of having your plan blow up in your face. The likelihood is low, but it could happen. Either way, great comment by which I totally agree with. I’m not scared. But more people should be.
Well said! 🙂
My husband and I retired in 2013 and have been very lucky with this market . Worried about a bear market coming so i have converted to cash or bonds the projected amount for all of our future withdrawals needed in our retirement. This leaves our portfolio at a 60|40 split . We are lucky to have pensions and ss in the future and only need dividends from our portfolio after those kick in.. I feel much better with this strategy, just have to see if i can stomach a 100% equity portfolio when we only need dividends. Thanks so much for your thoroughly researched and well written articles. Keep them coming .
You look like you’re in good shape then. Lots of folks with a two-stage retirement problem (pre-ss, vs. post-ss) underestimate the benefit that ss and pensions give you. If the time to SS is short, you can go way higher than 4% initially.
Agreed, we are probably averaging between 5% and 6% for the next 8 years and then hover between 0% and 1% for the rest. I used your wonderful spreadsheet to help confirm the SWR. Also living in Mexico helps. Our budget is 80% discretionary, so we can rein in spending quickly if needed. Thanks for all your insites.
Glad this was useful! Good luck! 🙂
Such a great article ERN. Envious of your research!
Many market commentators are saying this may be more of a prolonged fall then a full on crash. Somewhat scarier, as you said, is how long a market takes to react and in which markets they do!
Thanks!
A prolonged fall is worse than a 2008/9-style crash. Good point.
Excellent(if a little terrifying) article!
But, what do you think about the possibility that we are moving away from crashes and major bears by learning from past mistakes?
e.g.
♦The crash of 1929 was-in part- due to peoples ability to buy stocks on margin with 10x leverage- something nearly impossible now.
♦The crash of 1987 led to the introduction of breakers to prevent another Black Monday by shutting down trading for a period of time after a significant drop.
♦The Dot com crash had a very top heavy S&P 500 with too much concentrated in a single industry based on new tech. -Currently the spread across sectors isn’t nearly as bad even with FAANG (I think?)
2008 crash had a lot to do with a lack of bank regulations which have since been changed.
On top of those, the US has a current president who views increases in the Market as personal achievements and will seemingly do anything to keep it up. Sure black swans are always possible, but maybe as we go through them we de-risk by not repeating those particular ones.
All good points. I certainly hope that the next recession will not be as bad as past events. But nature has a way to throw things at us that are totally unexpected. In hindsight it all seems so obvious and inevitable but nobody had a clue right before the recession.
But I sure hope you’re right! 🙂
The next crisis will be triggered by something different than the last ones, and it’s not obvious what that will be (if it were, everyone would hedge against it and the crisis would not happen). If I had to guess, I would think it has something to do with >trillion $ in negative yielding sovereign bonds which probably represent the biggest bubble the world has ever seen. But we won’t know how the crisis looks like, and what the catalyst was, until after it happened.
Well said. I’ve heard all sorts of “scenarios” but once it happens we will still be surprised! But I agree: it will have something to do with the <0% yielding debt... :)
I have $185K set aside for college costs that I will start tapping in 2 years and continue to tap for a total of 6 years from that point (2 kids). I don’t have that money invested in equities as that’s too risky for me. My plan was if a Bear happens in the next couple years, dump half that money into equities and ride it up which could fund more of my kids education. This article makes me re-think somewhat. A flat market from the point I jump in won’t hurt me. I guess my biggest risk is a 20% drop followed by another large drop after I’ve “bought in” to the market. For that pool of money I guess I’d want the bear market to happen soon so my money has time to grow….
Yeah, good point! The dangers of market timing.
Consider putting maybe a fraction into the market if the bear market happns early enough and you still got 6 years of college expenses ahead of you.
There is no magic substitute for having more money they you need.
One should not retire at the minimum required. Your personal inflation may be higher than CPI. There may be a bear market. Whatever else may happen.
You need more money and/or a way realistic way to spend less money. Alternatively, one can continue earning money – much like you do. Live off the “side hustle” income, while growing the nest egg.
Very true! The personal CPI vs. reported CPI is also on my mind. Some people argue that then can grow their expenses at CPI minus x. I’m afraid mine will grow at CPI+x. Very good point!
Just to focus on one part of the article – you explain the problem with the bucket idea, which of course sounds good in theory until you sit down and try and figure out an algorithm of when to fill it up again and falls apart.
Using your idea of filling it up immediately doesn’t make any sense (and is a good example of why the idea doesn’t work).
One way I’ve read that some people deal with this is a bond ladder.
They will get maybe 5 term deposits, one for a duration of 1,2,3,4,5 years each with 1yr worth of living expenses.
With the 4% rule, this is 20% of your portfolio.
80/20 is incredibly aggressive for a retiree, so if we go with a 60/40 portfolio just as an example, you would end up with 60% equities, 20% bond fund, 20% in bond ladder.
In a bear market your equities drop and your 20% bond fund would usually hold up pretty well (assuming govt bonds), and as your 1yr bond rolled off to use for living expenses for the next year, you would create a new 5yr bond from whatever was most overweight with your equities and bond fund, so you would generally end up taking from your bond fund and leaving your equities alone. You would also be taking from your bond fund to rebalance “into” equities since your allocation would change when equities dropped, so you would also be buying equities cheaper. Again this last part is an abandonment of the bucket strategy and is using a percentage allocation instead which makes more sense.
The simpler version to having a bond or TD ladder is to just have a short term bond fund with average duration of the whole 1-5 years, so say a 3 year bond fund.
I’m not advocating for this idea, just pointing out that it would have been nice if you went into it a bit more rather than stopping at the assumption that you would re-fill your bucket immediately because if that is the case, then certainly the bucket strategy is just dumb (which it pretty much is since there is no way that can be properly argued that makes sense).
The devil is in the details. Again, at some point you have to replenish the bond ladder. Whether you do 5 years (80% stocks 20% ladder) or 10 years (60% stocks 40% ladder), this dynamic allocation however you want to structure it (Cash bucket, Glidepath, McClung’s Prime harvesting) will only slightly alleviate Sequence Risk. It’s like squeezing a balloon: you think you solve one problem but you create another somewhere else…
Yes I agree. I probably did not explain it well.
Say you had a 60/40 allocation where half or your 40 was in a bond ladder and the other half in a bond “fund”.
60% stocks
20% bond fund
20% bond ladder (5 year ladder)
When stocks are down 25% say and your next year ladder rolls off, you are left with
45 stocks (25% depleted)
20 bond fund
16 bond ladder (taken 1 year out to live off)
Your stocks to “total” bonds are now 45:36
To shore it up to 60/40, you want to change it to 48:39
Which means you are
• Moving from your bond fund to bond ladder: 4
• Moving from your bond fund to equities: 3
So basically you have replenished your bond ladder from your bond fund while equities are low.
You have also bought equities while they are cheaper.
Essentially, you are topping up your ladder from whichever is higher from your equities or bond fund.
Bleh I can’t explain it any more if that isn’t clear.
Basically, yes if you use only a bucket, it makes no sense. I agree. Which is kind of why I wondered why you even put it in where you replenish directly from equities.
I would have been curious to see the outcome if you had a bond fund to make up the rest of your bond allocation and instead replenished from whichever was higher from your stocks and bonds.
Yeah, I know where you are going. You do some active S/B allocation along the way. If you look at SWR Series Part 25, I did something similar to that: 80/20 portfolio PLUS Cash bucket. So, so live off your cash bucket and do the 80/20 portfolio with reallocation B->S the way you explained.
But, alas, that also will only alleviate Sequence Risk.
Amazing post as always Big ERN.
But I have a complaint: your analysis cover a 100% stock (S&P500 portfolio). Isn’t a more balanced portfolio (60/40 or 75/25 rebalanced periodically) an edge against such scenarios? I wonder how long it would have took to a 60/40 portfolio (rebalanced periodically) to get above water. Do you have some data?
Good point!
Yeah, this only referred to the stock portion for now. But my point is that the length of the bear market will spell trouble even for an 80/20 or 60/40 portfolio. Imagine you start with 60/40, wipe out your entire bond portfolio after 10 years. Then the stock portion has to, again, grow about 5% real (100/60)^0.1-1=5.24% p.a. in real terms to make up for the exhausted bond portfolio. But in some of the bear markets it takes longer than 10Y for the market to catch up to CPI+5%!
Also, I have done simulations for a 80/20 portfolio plus cash bucket on top of that (!) in SWR Part 25. It helps somewhat. But it’s no panacea.
Great post, this is a follow-up question. Would like to see the charts for a 60 Equity/40 Bond (and 80/20) allocation with annual rebalancing to maintain static original allocation. You never “wipe out” the bond allocation since you are rebalancing. And for the Bond allocation, I suggest using the AGG rather than Treasuries, even if the AGG only started in the early 70’s with Lehman Brothers. (I suspect most DIY’s use the AGG to keep their portfolios simple especially given its out performance over 5 yr Treasuries the last 10 years.)
I don’t do annual rebalancing. It’s a pain in the butt to simulate this.
Also caution: AGG has higher yield but also higher correlation with the stock market. Keep in mind my posts Part 31-33 in the SWR Series! Higher yield means higher SoRR for fixed income!
Great article! And you made me feel even better about having moved the whole megillah into Vanguard’s 2015 Target Retirement Fund a couple of weeks ago. We’re comfortably retired at 70 years old, so it makes no sense to court unnecessary risk. And it’s worth a couple of bucks on a thousand to have Vanguard do the rebalancing. My days are too filled with pickleball, bridge, pinochle, travel, and walks in the woods to clutter up with financial clickwork once a quarter.
Thanks for sharing your experience.
I’m not a huge fan of target date funds. They keep lowering the equity share through retirement, which is the opposite of what is optimal from a Sequence Risk point of view! 🙂
You never cease to deliver thought provoking and informative posts. It’s not hyperbole to say you’re the best FIRE blogger out there. No BS, no selling stuff, just the numbers, actionable strategies, and a touch of humour.
This article has scared the bajeezus out of me. Good timing being Halloween today in Australia.
Despite being scared, is it the right take away that no change is required for the accumulation phase? Just keep saving and investing in equities/property/derivatives/growth assets? Being scared and staying on the sidelines is not the way to go.
The actionable parts would be things you’ve mentioned in part articles. Over-accumulating and conservative WR, along with glide path?
Am I missing any other takeaways from the post?
I would still stick to my aggressive equity strategy if I were 5+ years away from retirement. But closer to retirement I’d look for some alternatives: option writing, real estate, etc. 🙂
Yes, plan to to accumulate a bit more, something that would have survived past market crashes. WHich is not that much more than 25x, by the way!
Great post. It pains me to see the number of people in the FI space chanting 100% VTSAX even after someone has enough money to retire. Derisking the portfolio is critically important after getting that big pot of money, hanging on to a high beta in a portfolio can crush an investor.
Yeah, I hope not too many people in the FIRE community do the 100% equities. If you have blogging/podcasting income you might, because the income is almost like a bond portfolio. So, even the folks screaming the loudest about 100% equities have an implicit bond portion…
Great post!
I believe to quote W. Buffet is appropriate in this context also:
“Only when the tide goes out do you discover who’s been swimming naked.”
Haha, that’s one of my favorite quotes by him! 🙂
Hei There,
thank you for that great content. Well, I am afraid of bear markets. And guess what, your calculation (Maybe because it’s very individuel) has not included taxes and fees.
…And don’t forget, you have to pay on the inflation taxes. The Governments takes taxes on nominal profits not on real profts !!!!!!!
Yeah, your 4% gross withdrawal is still taxed. Though lots of retirees can structure their withdrawals to keep taxes minimal. Also fees have come down a lot 0% to maybe 0.015% with Fidelity.
Great analysis BUT THERE’S MORE! You assume a constant inflation adjusted SORR of 40K/yr but RMD adds additional stress in that it’s progressive, so by year 10 the government is forcing a WR of 5.3% on deferred funds rising to 9% by age 90. You of course don’t have to spend 9% but you do have to pay taxes on 9% and taxes are progressive. So RMD is progressive and taxes are progressive and that causes a progression in SORR. In addition if you’re MFJ and a spouse dies the tax rates is then calculated on a single income and can progress taxes by as much as 2 brackets. For me it was a great motivator to Roth convert and to do that as cheaply as possible.
I did more work on SORR vs when it occurs in the period of retirement and SORR is effective for half of the retirement period NOT 10 YEARS. In a 50 year retirement it’s operative for 25 or so years. The 10 years bandied about comes from a paper that describes 15 years as the operative length of effect in a 30 year retirement. In terms of risk mitigation the first 5 years of the 15 predates actual retirement and the last 10 are into early retirement and the period is based on a 30 year retirement. If you were to graph lifetime risk vs time you would come up with an inverted curve starting low with accumulation reaching a peak at retirement and then a gradual reduction in risk as you approach death, but the curve is asymmetric with a fairly explosive increase in risk during accumulation and a slowly decaying in risk during retirement, assuming enough money in the portfolio
Your analysis describes my contention that a 30 year retirement with only 4x living expenses is actually very leveraged, more leveraged than typically assumed. This is shown by extended recovery periods post recession to get back to zero (if you ever do) plus the need to make a greater return than what the “formula” dictates. I’ve been reading a book by Mandelbrot and I am pretty well persuaded by fat tails and the notion of market turbulence which would further exacerbate SORR. Then there are the macro economic concerns of boomer retirement as a direct threat to GDP. You and I are part of that threat since we necessarily spend less than we did when employed. Lowered GDP is a increased SORR equivalent. So the issue becomes
1 length of retirement
2 WR
3 a weak economy based on the macro environment
4 portfolio over leverage required to keep the portfolio above water
5 increased tax and RMD progressiveness
6 Inflation
Short retirements with low WR amd funds primarily NOT in IRA accounts have the least risk. Portfolio’s with assets in different non correlated classes also reduce SORR because you will typically have at least 1 asset class that high during any period and the rule is sell high. Dalio and Browne suggest such portfolios. I’m not a fan of bucket theory portfolio’s which constantly shifts high risk assets to low risk over the course of retirement. I am a fan of having what I call a fuse portfolio, which is a second portfolio of a different risk about 4 or 5 years of need, that you use once, and then do not refill. The point of this is to effectively re-sequence bad SORR. The fault of SORR is it forces you to sell low. By having a second smaller lower risk portfolio you have something to spend that is not high risk money. The second portfolio is something like 15/85 stocks/bonds. If the market crashes by 50%, 15% stocks drops by 7.5% leaving you with 92.5% in the portfolio to use to buy hamburgers while the high risk portfolio recovers. If your crash is 10 years into retirement this lower risk portfolio has seen good gains as a 15/85 typically returns 5%/yr. By spending low risk money you effectively re-sequence the risk to a better long term outcome. If you look at FIREcalc there is a graph of a 1973 retirement, a 1974 retirement and a 1975 retirement. I you re-sequence a 1973 retirement by spending low risk money and leaving high risk money closed to withdrawal and move to a 1975 retirement the outcome dramatically improves. SORR is about a sequence so you have to game out the results of actual sequences, not just smoothed averages. I sleep just fine because I understand the real nature of the risk and I’ve hedged against it. Tnx as always for a thought provoking narrative
I just got my SS estimate in the email last night. It got me to thinking about using SS as a timing technique. Let’s say you retire at 64 and go on medicare at 65. Your intention is to pull the SS trigger at 70 along with RMD. You begin Roth conversions with the idea of partially emptying the TIRA while living off cash till age 70. You could have a big cash pile to last 6 years but alternatively if the market crashes big (50%) you could simply pull the SS trigger a bit early. You would forgo the guaranteed 8% excess return on the annuity but you would significantly reduce the WR on the portfolio and therefore decrease SORR risk during recovery as well as shorten the time tp get back to zero since the government is now paying a portion or all of needed income. At 70 you have about 20 years to live and so a decade long recovery period is not trivial in terms of risk management. If you cut the recovery period in half (say 5 years) through WR reduction when SOR risk is at it’s peak, with each passing year your portfolio becomes more and more bullet proof in an accelerating fashion.
By waiting till 70 your SS 25% greater so 3K/mo becomes 3.75K/mo at 70. Pulling the trigger 2 years early would yield 3.45K/mo or . If your WR is 4% on 1M (40k), pulling the SS trigger at 68 would give you income of 41,400/yr which is a net 0.035% savings on WR that means your WR is actually negative not positive. In addition SS is tax advantaged so you would get taxed on only 35,190 of income further boosting the yearly income 41,400 MFJ in FL is a tax of federal 1610. On 35,190 you pay $989. What this means is by pulling the SS trigger 2 years early you effectively close the portfolio to SORR while it recovers, and you can still Roth convert. If the TIRA is down from a crash, the Roth conversion is much more efficient. Stocks are property and Roth conversion of a depressed property value means you can transfer more of that property from a TIRA to a Roth for the same tax dollar. This means if you have 1000 shares of AMZN to transfer and the share price is $1000 you would be taxed on 1M ordinary income. If AMZN drops to $500 you get taxed on 500K ordinary income for moving the same 1000 shares. 1M is 300K taxes, 500K is 117K taxes for moving the same amount of property, a 182K savings in taxes for the Roth conversion. Do not miss this opportunity when the calamity comes. A 182K tax savings pays for 4.5 years of retirement @ 40K/yr. A free almost 5 years of retirement is a hell of a return on a 30 year retirement. People say you can’t time the market, they’d be wrong. If you look at the market from a return perspective it’s hard to time. If you look at the market from a risk perspective then you can game out all kinds of savings by choosing the path of least risk. I bet nobody in bogglehead land ever told you the best time to Roth convert is in the middle of a 50% recession.
Are you afraid of the Bear? Hell no! The Bear is a possible relative opportunity if analyzed correctly.
A good SS estimator for married or single is
https://smartasset.com/retirement/social-security-calculator#3KtwjEZpZ4
A good tax estimator is
http://taxplancalculator.com/calc
Very good points. I am doing a case study (to be published Friday) to talk about a sitiation just like this one: Wait until age 70 and fund all expenses out of taxable savings for 6 years. Keep taxable income low to max Roth conversions and then ive off SS and Roth IRA post age 70. No RMDs, very low taxes in the 12% bracket.
And if the bear happens in the next 6 years the Roth conversions become even cheaper! Perfect hedge! 🙂
Thanks Gasem! All really great points!
The death of a spouse is a concern because of the tax brackets and deductions going down by 50%, while the expenses will go down by much less than 50%. Well as long as the expenses still go down somewhat, it will offset the tax burden. But yes, I agree, the tax issue is a concern. We will definitely not assume that after the death of a spouse our pre-tax withdrawals will go down!
All good points on the SoRR. I never like to pin down a exact time frame. If one has a 50-year retirement horizon, SoRR can be a problem during years 1-40. If the portfolio only moves sideways in years 1-30 and then we have a bear market in year 31-33 you’ll still have a problem.
I like the idea of a fuse portfolio (if I understand it correctly). Reminds me of the cash bucket (see SWR Series Part 25) that you keep around but won’t replenish. It will take care of SoRR issues early in retirement (but will have trouble keeping you afloat in a 1960s-1980s scenario).
This is the kind of post that still keeps me thinking that I want 33 times my expenses saved up before retiring. I know that may be viewed as too conservative but when I stop working, I want to stop working.
33x seems overly conservative. somewhere between 28-30 will be safe even in the worst historical episodes. 🙂
Is there a mistake in your Table 1 (S&P 500 Price Index) worst 5, benign 5 calculations?
Re my earlier comment – I have just noticed the note under your Table.
Please can you explain why do you not consider 2007 amongst the worst 5?
Also, I assume the note should say 1937 rather than 1939.
Or, is it that your definition of “worst” is based on “duration of bear” rather than either “drawdown” or “months to full recovery”?
Yes, based on the duration. 2007 wasn’t that bad of a bear market from a SoRR perspective. I added a note to the table to make sure that’s properly conveyed. Sorry about the confusion! 🙂
There are rather a lot of Q’s above about de-risking, mitigation options, etc.
The fundamental issue, as I see it, is that the so-called “safe withdrawal rate” is unknown, and unknowable, in advance.
Hence, there will almost always be a non-zero probability of failure whatever “tweaking” you do.
If you cannot live with this then you may be taking the wrong approach for you!
An alternative approach, similar to that described by “RETIRED” above, does however exist.
This approach is recommended by life-cycle economics, and is sometimes referred to as safety first.
See, for example, this recent article http://www.theretirementcafe.com/2019/09/the-prevalent-but-problematic.html
Yeah, I’ve heard of the Zvi Bodie approach. Unfortunately, due to low yields, the “safety-first” approach yields you ay too little money, especially over a long horizon (50+ years), so you might as well try your luck with a low WR that would have survived the Great Depression and the 1970s.
I hear you.
However, it is not really a question of approach A or approach B but rather how much of A and how much of B? See for example, http://www.theretirementcafe.com/2018/01/unraveling-retirement-strategies-floor.html or, indeed, your own posts about the positive impact of SS and pensions.
As it happens, your latest post is a great example of such a case. I would be tempted to describe your Becky and Stephen scenario of yesterday (8th November 2019) as Floor and Upside. And, assuming they actioned something akin to your CD ladder suggestion prior to their SS kicking in, I would then view their plan as around 70% Floored for at least 25 years with their home held in reserve for a possible future move into an assisted living community.
Yup! That’s an excellent way of interpreting that case study! 🙂
Yeah, good point! I like the TIPS ladder approach!
ERN, thank you for another fun post. There are two blogs I would like to share with your readers:
https://portfoliocharts.com/portfolio/my-portfolio/ – very nice visuals but data only goes back to the 70s. You can play with the tool to hone your asset allocation to make it recession proof or at least less sensitive… but then again it is always about fear vs greed and greed always wins in the long term.
https://retireearlyhomepage.com/chronidx.html – this blog is by John P. Greaney. He successfully retired early in 1996. He details his journey and reader can see what he did or didn’t do during the last few recessions. Interestingly enough he increased his spending during the last recession to take advantage of all the travel and real estate deals that were available when no one else was buying.
Thanks for the links. Knew about both already.
Use real caution with the portfoliocharts.com WR simulator. It misses the historically two worst events (1929, 1965-1968)!
Greaney’s blog is a gem!
This site has a huge amount to offer in spite of its rather dated appearance.
He may actually be “the best-kept secret in the history of early retirement”.
For a fuller description of his contributions see e.g. https://forum.earlyretirementextreme.com/viewtopic.php?t=10436 And do not miss his 2001 paper “Drive Your Financial Advisor’s Porsche and Retire Before 40 — The simple arithmetic of saving and investing.”
ERN, thought provoking analysis as always.
Something that may be worth consideration. An allocation to cash need not be solely for drawing down income during an equities downturn; if the downturn is sharp and deep, and you have multi-year cash holdings, some cash can be used to capture the buying opportunity and then the recovery will be amplified.
Whatever strategy is settled on for handling a poor sequence of returns, it is probably best deciding exactly what it will be, at what market fall percentage you will buy and how much, and do this thought process before you start drawdown, with a clear head. Because if it happens and you have no plan, emotion will cloud judgement.
I agree. A glidepath would do the same thing: shift assets out of bonds into stocks over time. That would have the sane flavor as your suggestion. I’ve written about that in SWR Series 19/20. Again, it’s best to do this in systematic way to get the emotions out of the game.
Hi Big ERN – Great article. Do you have a post with your asset allocation? I would be interested what you would recommend for the accumulation phase versus your current retirement phase. If you have an article on this, please link to it. Currently we are early in the accumulation phase and have 80% US stock 10% international stock and 10% bonds, but we also have a 280 K mortgage. I might rethink the bonds… Thanks so much
I wouldn’t stress out so much over where you invest in stocks. US-only or US+International. Whatever makes you comfortable.
10% bonds: Some people need that to calm their nerves. While accumulating I mostly had 100% stocks. Certainly, while you still have between 0 and 15 or even 20x annual expenses there’s no real need for bonds yet.
And during that same period, there’s no need to accelerate the mortgage payments yet. Use that leverage tho build assets faster! 🙂
This is a great post thanks. While there are some great practices attached to the “FIRE” community, you don’t have to dig deep to see that there is (at times and in many places) very little understanding of stock markets and risk, in the accumulation and decumulation stages. This can be used as one of the great go-to pages.
Yes the killer is that no returns in real dollar terms for over a decade. I talked to too many US retirees in the early 2000s who had their dreams smashed, and back to work they went.
That said, simple worked in the last 2 recessions. I penned on Seeking Alpha how retirees in Vanguard’s simple managed funds breezed through the last 2 recessions with 4% plus spend rates, inflation adjusted.
Core US sensible large cap dividend payers plus some International stocks plus some bonds. VWINX, VWELX. Given low bond yields of today investors might slant toward a more generous stock allocation.
Also as Big ERN points out, you need to de-risk well before the retirement start date. Those who are doing ‘quick FIRE’ and are piling everything into stocks are playing with FIRE. So far, they’ve only got lucky, thanks to the longest of bull runs.
Of course a real market correction and a recession will throw buckets of water on most of this. I’d guess (from observations) that few will be truly prepared.
Again, great post.
Thanks, Dale
Very good points! I also think that stocks are now TINA (ther is no alternative), in light of all asset classes looking so expensive. But I certainly wouldn’t go 100% stocks either! 🙂
WHat’s the link for the Seeking Alpha article? Feel free to share here!
I’ve never seen any finance blog discuss this idea, but I try to treat my gains from the last 2-3 years (on a rolling basis) as ephemeral. Those are when bubble valuations are reached at end of bull markets.
The data are quite compelling. Here’s from eyeballing U.S. data (monthly, with dividends and inflation) since 1926 and since 1870, from: https://www.advisorperspectives.com/dshort/updates/2019/10/07/the-four-totally-bad-bear-recoveries-where-is-todays-market and https://www.advisorperspectives.com/dshort/updates/2019/11/07/the-latest-look-at-the-total-return-roller-coaster
* All past U.S. bear markets took only 1.5 years to recover to their level 2-3 years before the peak (even the Great Depression)!
* 2-3 years before the peak, the market in most cases was was only -25% lower (though the 1929 top went much higher).
* The Great Depression was really only approximately a -50% crash (not -85%) if you don’t count the crazy 130% gains during the 2 years before the top.
* So after ~2 years of bear market to trough, and ~1.5 years of bull market after trough, historically speaking, the market got back to where it was at 2-3 years before the top–usually -25% below its high.
(This is mainly an exercise in managing emotions (fear), as your withdrawal strategies were decided beforehand based on SORR assumptions. And remember to check your portfolio only once a month!)
Your portfolio doesn’t deserve to keep the last 2-3 years of returns until they get “baked in” by the passage of time. If the past is a guide, we need another year to bake in the big 22% gains of 2017!
Good point. A similar, more systematic way would be to look at the CAPE. If the market rallies without earnings moving much, you should ignore the recent gains.
But then again, if gains are supported by strong earnings growth, you probably should take your stock rally into account. 🙂
Also thanks for the link. I loved the picture with the 4 bear markets: the recent one was the most tame after CPI. The other three would have been devastating for retirees. Exactly what I tried to convey here with the post! 🙂
I totally share your concerns. When you read today’s FIRE blogs, it seems like everybody is just swimming in a hot calm sea. To say you can tolerate a bear market and to actually stay put and keep investing thru one when you see every dime you put in disappear the next day and more then you won’t be so sure about things like you are now. Believe me I’ve been thru one and it’s scary as hell.
PS: a shout out from Brazilian FIRE community !
Thanks! And a big Obrigado back to beautiful Brazil. Visited a few times many years ago!
Yup! A bear market always looks fine in hindsight. But sticking with your investments through the cycle is not that easy. And it becomes even more difficult when in retirement!
Fantastic write up Ern! This is definitely inline with what Ray Dalio says on how we’re entering a paradigm shift similar to what happened during the great depression. Although he sees it more likely than you that we will be entering a recession.
I’ve done a write up on gold and how it can help through a market crashes.
https://encoreyourlife.com/why-the-rich-are-buying-gold-to-get-even-richer/
Thanks!
I’m not too much of a gold bug. But it seems to have helped somewhat during all the major market crashes. Maybe do it through futures? I would hate to give up much space in my portfolio to a zero-yield asset. 🙂
Yup, that seems to be a major sticking point, the lack of yield.
In addition to holding gold, I’ve also been collecting yield through options premiums from selling puts in the etf’s GLD and GDX. Because fo my long term bullish view I’m quite stress free and consider it very low risk. I sell the puts with a strike around market price (when I do the selling) with an expiration of about one month out. If during expiry the price has dipped below the strike, I just roll it out about two weeks or so (as long as the price hasn’t dipped too much) and keep rolling if its in the money. I’ve been assigned early a couple times when doing this but I just sell out of the position and re-establish the trade by selling the put again at the same strike price. I’ve found this has been working quite well but would only suggest it if having a bullish view in the long run.
I like the idea of the gold short put options: It gets over the no-yield issue! Nice suggestion!
not gonna lie, this article made me nervous. Retired 3 years ago. Portfolio was 1 M and withdrawals the last 3 years have averaged 42K a year. Portfolio now is 1.3 million. Im 17 years away from SS at age 70. I’m at 95% stock. What should I do?
Well, with 1.3m in assets and $42k annual withdrawals your WR is only 3.23%. And you have to bridge only 17 years.
Looks like you’re in excellent shape.
In hindsight you also did really well with a high stock allocation. Going forward, I’d probably shift a little bit more into bonds. Why risk so much when you already won the game?
thx ER. You know what it is? I’ve just always had a very high stock allocation and just mentally parting with that I struggle with. Then, when I do the numbers, even if we do have a 40-50% drop ( which as you know is very rare) I’m still okay when I take the reduced amount and run it on FIRECALC, etc and I also can drop my spending a bit. Taking 10% and putting into bonds would maybe just stabilize things a bit in the event of a crash I guess.
Me too. But that’s OK in the accumulation phase. Not so much when withdrawing! 🙂
It’s definitely scary because most people didn’t go through a bear market yet. They think that everything will be just fine. But at the end of the day, if people are not prepare to face a bear market, shouldn’t they reconsider their investment strategy?
I think your blog post is well written and easy to understand. Thanks for putting the time to write it!
Thanks for the feedback!
It’s an “ignorance is bliss” kinda deal. Once the bear market comes people will adjust. 🙂