Sequence of returns risk and early retirement

[thrifty++]

I have been thinking a bit about sequence of returns risk lately.

Most of the people I have noticed on here have early retired from 2010 onward and during this time have enjoyed great returns on their cash. However I dont think its too speculative to predict that we are probably in for some degree of a downturn in the next couple of years since we have had such a long bull run and a new debt binge bringing debt to an all time high. So this makes me a little nervous about early retiring soon and having my assets decimated immediately at the beginning of the retirement = very bad sequence of returns.

I am interested to hear from anyone who has early retired say in 2006/2007/2008? And how did that go for you? Especially those who retired on smaller sums eg say $150k. I think Jacob might be in this category? Anyone?

Also interested in thoughts on sequence of returns risk generally.

[theanimal]

Early retirement now has a great series on this:earlyretirementnow.com/tag/sequence-of-return-risk/

It's been mentioned on the forum before but I'm not sure there is a dedicated thread.

[jacob]

I called it quits (told my boss that I was pursuing "other interests") just about a week before the market bottomed out in 2009. My NW had previously been higher and I knew that valuations around the bottom was approaching the "historical average" (not below average, mind you, just around the 100% level. Markets are currently at 220%+). More importantly, I had a back-up in the form of a side-income: http://earlyretirementextreme.com/my-4- ... -week.html

Initially, since early 2008, I had been switching into blue blue-stuff ( http://earlyretirementextreme.com/time- ... ssion.html ) so I took less damage than the market. As banks and funds started folding and liquidating to cover their debts, value screens (I was more of a value investor back then), started to offer real choices so I switched into those as the market kept losing 20% 30% 40% and selling indiscriminately ... so the recovery was very good to me. After a couple of years, I had doubled my NW(*). If you pull up the graph now a decade later, it seems like a blip. Living through it day by day, it was harder. Consider that WWII only lasted 6 years. However, I'm sure that on a day-by-day basis it felt like forever.

(*) Please don't bother to infer details about returns, etc. using some back-of-the-envelope calculation. I split all income and all realized capital gains with DW 50/50 and historically I've earned more than her (all things included) in almost all the 10+ years we've been together except 2009-2010.

However, I should also add that DW was working full-time, so there was an ultimate backstop in case it was needed. Our budget is low enough that we have more than we know how to spend on just one minimum wage income at all times, so we don't worry too much. One of us would be able to do this alone or in combination with different kinds of income. This is why ERE is more resilient than standard FIRE.

In 2012 I started a job in finance which made ~twice my salary as an academic researcher. Indeed, almost as much as an entry level software engineer with a bachelor degree. How crazy is that? ... I worked at that for almost 4 years. However, because we moved, DW quit her job and started another career from scratch (lost her passion for sampling ground water around gas stations) and was making <20k for a few years at the same time, so it evened out (sharing, remember) However, again ... since we've been spending <$14k/year ... any "normal salary" is ridiculously over/out-sized compared to what we need and want at ERE HQ. So NW kept increasing.

In 2014 we exchanged part of the portfolio for owning a house instead of renting. QE was ending. I knew that market returns strongly correlated with the Fed's various QE schemes and RE was undervalued. Also, running ownership costs were about half of rent, so it was a no-brainer. Currently we spend $11k/year for the two of us.

We currently have 95x annual expenses saved between the two of us (do the math). DW is still working and has the fall-back skill to pick up seasonal work anywhere within the US and make enough from January to April to support the both of us for the entire year should we lose ALL assets. I quit my finance work in 2015, but still I have this website and the book royalties but I like to think I have the fall-back skill to lift heavy stuff and make some high four-figures (all we need) per year somewhere, PPP-adjusted, if it comes to that [that is, total financial collapse] AND if people no longer care to read about what I have to say, SIMULTANEOUSLY.

In conclusion, we're living ERE the way I originally envisioned it prioritizing production over consumption and always trying to add value. Decisions along the way were made with the goal of developing the renaissance lifestyle further. (We've also been fixing up a fixer-upper and grown and preserved about a month's worth of food/year out of our garden although I for one realized I don't like to do this. But at least I can do it if I have to.)

I think we're now in a position to fight at least 4 fronts at the same time. There's just so many ways we can make enough income to cover our needs and wants. So overall a strong defensive position.

That said, I'd be super-worried about the young jedis who started investing around the 2009+ era... whether it was leveraged RE or total market ... because I worry they've come to believe that this [return trajectory] is the norm; especially if they permanently bought into the "simple" paradigm. Also, because the 4% seems to be treated as dogma regardless of the fact that we live in outlier-space now. Combine this with those who haven't bothered to reduce their spending enough so ANY income will do (including burger flipper) and I foresee the potential for significant pain for the 'simple optimist'-demographic. Just goes to show that the meta life-hack strategy of "picking what you want and ignoring the rest" is ... crazy reckless. 'Complex-realist' should be alright though.

For those who don't want to adopt the full ERE strategy, I wouldn't be comfortable FIRE'ing with anything less than at least a 1.5% rule at present valuations. To me .. if you worry about sequence-of-returns, it's a strong indicator that you either don't have enough diversification in terms of other value-add or that you haven't saved nearly enough. IOW ... you can't or shouldn't fix cashflow problems in reality with fancy math gymnastics(*).

(*) Mathematically speaking, SOR is being in the precarious position of having to worry about the worst lines in an MC simulation.

[IlliniDave]

Generally speaking it is my number one "worry" right now. I'm on the cusp of pulling the plug and so on the cusp of being at my most vulnerable. So of course volatility would return right now.

Most likely I'll just delay cutting the umbilical. 15 months from now my retirement benefit jumps at which point I'll be in a five year window where it ramps up quickly from there. By ~3.5 years from my planned date (a little over 5 from now) I may not need to touch my stash at all (SWR = 0%, inf*X) unless inflation really spikes dramatically, and even then the need to start withdrawing would be years down the road (hyperinflation aside). I'm also dialing back my exposure to stocks, which is standard procedure in some investing schools heading into retirement. Lowering exposure to SoR risk is among the reasons for that. Maybe that's a long way of saying I'm totally overreacting.

The upside is that understanding the risk and knowing myself well enough to gauge my response should allow me to circumvent any disasters. The downside is that you all here might have to put up with me for a little longer.

[Mister Imperceptible]

95x annual expenses

****

That said, I'd be super-worried about the young jedi Millennials who started investing around the 2009+ era... whether it was leveraged RE or total market ... because I worry they've come to believe that this [return trajectory] is the norm; especially if they permanently bought into the "simple" paradigm. Also, because the 4% seems to be treated as dogma regardless of the fact that we live in outlier-space now. Combine this with those who haven't bothered to reduce their spending enough so ANY income will do (including burger flipper) and I foresee the potential for significant pain for the 'simple optimist'-demographic. Just goes to show that the meta life-hack strategy of "picking what you want and ignoring the rest" is ... crazy reckless. 'Complex-realist' should be alright though.

For those who don't want to adopt the full ERE strategy, I wouldn't be comfortable FIRE'ing with anything less than at least a 1.5% rule at present valuations. To me .. if you worry about sequence-of-returns, it's a strong indicator that you either don't have enough diversification in terms of other value-add or that you haven't saved nearly enough. IOW ... you can't or shouldn't fix cashflow problems in reality with fancy math gymnastics(*).

(*) Mathematically speaking, SOR is being in the precarious position of having to worry about the worst lines in an MC simulation.

Congratulations Jacob.

****

With all due respect and no malice whatsoever intended for late accumulators IlliniDave et al., this is actually encouraging for we early accumulators. I need to establish my own value screens. Cash is king!

[Lucky C]

To use round numbers, let's say if valuations were in the normal range you'd quit with a 4% SWR, but considering today's valuations you'd only quit when you got up to a 2% SWR. Then you can make a simple formula like SWR% = 64/CAPE so you can see what SWR you would want as valuations fluctuate.

Now the question is which will get you to your goal first: continuing to invest and riding the waves of the market, or holding cash and waiting for your savings to be at a comfortable level before you invest and quit?

If the bull market keeps going, you will have to get to 2% SWR or even lower as the market continues to rise - this will probably take several years whether you are invested or in cash, if you originally planned for a 4% SWR. If on the other hand we enter a bear market in the next couple of years, you could get closer to your goal much more quickly if you are in cash vs. if you are invested heavily in stocks. Since valuations are so high right now and the bull market has been going on for such a long time, it seems to me that there is a greater potential to benefit from being in cash if you are close to FI, since the alternative is to hope for even more extreme valuations / market melt-ups for some quick and easy gains.

I'm in a similar situation now where I would be comfortable quitting if market prices were average. To get to a 1.5-2% withdrawal, I would probably need to work another four or five years. But it seems to me such a high probability that we'll be in the next bear market at lower valuations before then, that I have high confidence that I will reach a safer FI level sooner by holding mostly cash for now. A market crash / start of a bear market is not all that unlikely this year, and that would be a dream come true to me.

P.S. Did you achieve your FI date much sooner than expected, compared to your initial FI date projection years ago, due to better than average market returns? If so, maybe you should have kept your projected FI date static all this time to not trick yourself into thinking an overvalued market would let you early retire even more extremer.

[thrifty++]

Interesting comments on this issue above.

I think it would be terrifying to quit your job and then suddenly have your NW reduce to half its previous value. Would be very bad luck.

[Tyler9000]

To use round numbers, let's say if valuations were in the normal range you'd quit with a 4% SWR, but considering today's valuations you'd only quit when you got up to a 2% SWR.

That's not how it works. The 4% SWR was calculated from the worst historical start date, not the average, so it already accounts for poor valuations. And more importantly, valuations really don't tell the whole story.

For reference, the single worst 30-year period (that set the 4% rule) for a traditional stock/bond retirement portfolio in the US for the last century started in 1966. Shiller P/E was about 24. In contrast, in Shiller P/E spiked to over 43 in 2000. But looking at my data, year 2000 retirees with a normal 60/40 portfolio are still on pace to have a 30-year SWR closer to 5% (I have a method for projecting that based on how the math works). Poor SWRs are a lot less about high valuations and more about stagnant long-term growth and high short-term volatility.

Beyond that, the single biggest shortcoming of almost all SWR analysis is a faulty assumption that there are only two possible index funds available for you to invest in. Break those artificial handcuffs and the 4% rule of thumb is actually quite low. Other portfolios never considered by those studies have easily sustained rates higher than that while never losing inflation-adjusted principal, which is the superior metric I personally recommend for very early retirees like you find here. My own portfolio has a measurable perpetual rate of about 5% that has remained remarkably steady through all types of economic conditions both good and bad.

Long story short, I think a lot of the angst about stock valuations affecting withdrawal rates is an artifact of value investors thinking about the problem based on their own paradigms. Think about it from a different perspective where there are more ingredients in the pantry than just stocks and a few intermediate bonds, and maybe you'll have better luck.

[thrifty++]

Long story short, I think a lot of the angst about stock valuations affecting withdrawal rates is an artifact of value investors thinking about the problem based on their own paradigms. Think about it from a different perspective where there are more ingredients in the pantry than just stocks and a few intermediate bonds, and maybe you'll have better luck.

What other ingredients are you thinking of?

[Tyler9000]

What other ingredients are you thinking of?

Small caps, international stocks, value stocks, long term treasuries, gold, commodities, REITs -- all sorts of things never accounted for in most retirement research that have positive effects on withdrawal rates.

Oh, and back to your concern about early losses you might just consider other portfolio options. Just as one example, the Permanent Portfolio lost only 4% in 2008 and has supported retirees quite well.

Here's a fun chart for thought:



You can find a guide to the portfolio names here. Note that not only do diverse portfolios have very different withdrawal rates, but there's also no noticeable correlation between withdrawal rate and average return. While that seems very counter-intuitive, also note that 4 of the top 5 portfolios on the list have no more than 40% stocks which effectively mitigates the possibility of market crashes killing portfolio performance. So personally I worry less about stock valuations and more about minimizing volatility of the portfolio as a whole, as that affects the numbers quite a bit.

[Tyler9000]

BTW, I should note that even though I study this subject a lot I still do not advocate pushing the limits of calculated SWRs. I see them more as a guideline of what NOT to do rather than a guarantee of success, so be smart about it and plan conservatively. I also completely agree with Jacob that retirement resiliency is about so much more than mathematical "safety" and the other aspects of ERE are just as important if not more. Still, if you'd have a higher SWR investing 100% in T-Bills or TIPS and dividing your portfolio by your expected lifespan maybe you're aiming a little low and should think about approaching the problem from a different perspective.

[ThisDinosaur]

First let me say that Tyler9000's site eliminated my concern about SOR. So thanks for that, Tyler.

One insight I'd add to what he said so far is that volatility is only your enemy if you can't buy the dip. If you have an income, cash, or an uncorelated asset, volatility is your friend because it means more buying opportunities.

The other thing I'd say is that a 3% SWR is historically conservative, and even in a >50% loss gives you a decade+ to figure out how to compensate. You can minimize your exposure to historically overvalued assets as well.

I'm a pessimist by nature, so there's plenty of other risks I think about. But SOR risk is manageable.

[Solvent]

I also worry about SOR. I am self-aware enough to realise that I started investing around 2009 - and things have been pretty damn easy since then. I think that however I project out my possible retirement/FI dates, I really ought to add at least two years to the timeline because I suspect there'll be at least one major market disruption in the intervening time. I have to remind myself that whatever headline figure I count as my NW right now, it's not unreasonable to consider the chance that it may be ~20% lower in 6 months' time rather than incessantly rising each month.

[IlliniDave]

BTW, I should note that even though I study this subject a lot I still do not advocate pushing the limits of calculated SWRs ...

Thanks for all the data. I'm not familiar with all those different portfolios, but I think the center-of-mass of mine is probably close to 3-fund (although I have some small/small value and EM tilt going on).

I tend to agree that back-tested SWRs shouldn't be relied on with complete faith. Maybe I've read too much of Dr. Bernstein's writing, but standing on the precipice I suddenly get the willies, haha. Another thing about back-tested SWR metrics is they assume perfect investor behavior. I think the average investor loses something like 2%/yr, and even though we're all Lake Woebegone residents here it's best to consider the possibility of imperfection. In addition the SWR math has predictably bounded spending needs as a premise. Neither perfect investing behavior nor perfectly bounded spending needs have been characteristics of my life. Although I compute an SWR metric, ultimately I don't base decisions on it.

Of course in my case looking at retiring at 55 with a more traditional arrangement, I'm really not an ERE-er. Anyone who a) plans to pursue paid employment/proprietorship of some form during retirement, and b) is ambivalent about (or against) leaving an estate/legacy should take much of what I say with a grain of salt. I score dying with less money than I retire with as a failure, which is a high bar.

[thrifty++]

@Tyler9000 - thanks for the insights. Thats an awesome graph. I might research those portfolios. Yes good point about Permanent Portfolio. I had not thought about it.

[Mister Imperceptible]

If you are heavily invested in stocks, and you are interested in the Permanent Portfolio, check out Tyler9000’s Golden Butterfly. It’s 40% equities. If I were ERE already I would implement it, if I didn’t have so many reservations about long-term bonds.

Tyler9000 didn’t bring it up because he is being modest.

[Lucky C]

@Tyler9000

The point of my post was that if you (a) believe that 4% is not safe in the current market environment, (b) believe that we are near the end of the current market cycle, and (c) are close to early retirement, you may want to consider moving some money to cash as you finish up your working years, since you don't expect too much gain from now until FIRE yet you are exposed to a fat tail risk. Of course my comment does not apply if you do not believe (a) or (b).

In my post I was not making any claims as to what is safe now vs. what is safe normally, just building on Jacob's idea that the safe withdrawal rate for a standard portfolio now should be more like 1.5%. I used 2% as an arbitrary example, but I don't recommend any particular withdrawal rate. There are so many variables for each individual to consider.

But since you brought up your recommended portfolios as evidence that 4% withdrawal rate is very safe, there are also some concerns with relying on those portfolios in this environment:
- For the US stock market, backtesting starting with a 1972 CAPE of 17 and ending today with a CAPE of 32 results in about +1.4% returns per year just due to valuation expansion alone. Your backtest captures several cyclical bull and bear markets but it is mostly one secular bull. If your backtest was 1936 -> 1982 (CAPE 17 -> 7) for example, the stock market components would not look so hot.
- Small cap alpha may be nonexistent in the future - backtests showing small cap outperformance may be too optimistic
- International stocks are at better valuations than US but historically not great valuations, and "when the US catches a cold, the world catches a fever"
- Gold backtest from 1972-current may be too optimistic - if gold is expected to match inflation long term, it may currently be overvalued, so real returns could be expected to be negative over the next decade or two
- Likewise backtesting bonds from the 70s on could be too optimistic, with expected real returns around 0% for the foreseeable future

Therefore while I believe a diversified portfolio like the Golden Butterfly would offer a nice low volatility ride, my personal view is that its expected return over the next decade will be around 0% (since all components would be returning close to 0%), so you could be down 40%+ at the 10 year mark, which could be a huge psychological barrier even if it ends up being followed by wonderful returns.

Please don't think I am hating on your site and your ideas with this comment. I have used the site several times and I think that it's great. However I am always thinking about how this whole FIRE thing can blow up in my face, so I am taking a conservative approach and currently I do not trust any buy and hold portfolio @ 4% withdrawal to last as long as I would like it to. I understand why Jacob is using 1.5% as his safe recommendation if someone plans on a simple buy and hold portfolio with no plans for future income streams - though I don't think many people on this forum plan on doing just that.

[Tyler9000]

No worries. We have very different paradigms for how to invest, so it's understandable we might have different outlooks. The best I can do is bring solid facts to the table, and they're all public for anyone to explore for themselves.

I do feel like there's a happy medium somewhere in between putting all of your faith in the calculated SWR for a certain portfolio and putting so little faith in any portfolio that you first save 67 years of expenses when the average lifespan today is about 78.

[bryan]

Timely as I am in the transition phase from accumulation -> FIRE. I just found out I am sitting on a 62% cash allocation thanks to needing to update my journal here

I seem to remember some nice discussion about sequence of return around the time @Tyler9000 released PortfolioCharts.com, either here or MMM forums.

Here is @Tyler9000 promising a nice feature dealing with pre-FIRE and post-FIRE portfolios (and perhaps the transition stage) which I don't think ever came about? Or maybe it did, but one still needs multiple tabs with the output of the first (accumulation phase: goal to quick working ASAP) feeding into the second (retirement phase: goal to not run out of money before you die content).

[Mister Imperceptible]

If you are convinced we are going to have a 0% real return across all major asset classes over the next decade because current valuations are too high, why not overweight cash (and perhaps gold also) so you can buy equities later? (YES, I’M TRYING TO TIME THE MARKET.)