Well, whattaya know… 7 years ago today-ish I walked out of my office building for the last time.
A lot has happened since that day. We traveled full-time for a couple of years before becoming parents and traveling parents… Jr has now been to 40+ countries. We started a couple of blogs, Winnie wrote a book, and we had maybe 5 minutes of fame when our story was shared throughout the media. We lived large, practiced intentional lifestyle inflation, and grew as people and as a family.
Thankfully, throughout all of this, our investment portfolio has also continued to grow. This is largely due to an extended bull market in US stocks, but also in small part due to blog and book income. (More is more.)
It’s been a good ride.
But… what if things didn’t go so well? What if the stock market crashed and the economy tanked? What if our hobbies never made a dime? What if it all went to hell?
What if it all went to hell?
No matter at what age, this is a good question for everyone to ask themselves before taking the retirement plunge. If we were unfortunate enough to retire in 1929 or 1965 or 2000, periods with strong economic headwinds… what would that look like?
Since retiring, even after withdrawing all of our living expenses, our portfolio has grown by about 50% adjusted for inflation (inflation over the 7 years is about 11%.) Had we started our retirement at a different time, but with the same inflation-adjusted withdrawals (spending the same amount), things would not look as rosy.
Using a 90% US stock / 10% US bond allocation (similar to our portfolio) with our actual spending, normalized to a $1 million portfolio, 7 years into retirement our portfolio value would be:
Start 2013: $1,540,000 (up 55% – better than our actual portfolio, International equities have reduced our actual portfolio growth)
Start 1965: $842,000 (down 16%)
Start 1929: $723,000 (down 25%)
Start 2000: $655,000 (down 35%)
Assuming no adjustments to spending (continuing to spend at today’s level going forward) and no Social Security or other income, we would be destined for failure, running out of funds in another 13 years or so. This chart makes that point clear.
Also shown in this chart:
1) how blog income has impacted our portfolio (starting to be significant)
2) impact if we didn’t spend less than 4% (relatively minor impact)
What’s a guy to do? I mean, besides not increasing spending through a falling portfolio.
Think (and Act) Different
Awhile back I shared how we have increased our spending as our portfolio has grown.
I think it interesting to consider how we might have behaved in a worse economic environment. It’s a good stoic practice.
The 1929 retirement cohort seems the most terrifying to me, as the stock market dropped about 75% over the first 4 years. (Over the same time, deflation dropped the price of everything by ~25%.) With a stock market plunge of this order, it doesn’t matter if you started spending 2% or 4%, you were still down over half after 4 years.
I can’t imagine that we would enjoy that experience, but let’s see how it may have impacted our early retirement. The following estimates are based on keeping a constant quality of life, just in a different location –> we spent $2k/month in Mexico/Thailand, $4k/month in Taiwan, $8k/month in Europe/Japan.
Year 1, 2013 / 1929, Actual spending = $38,966 (See the full 2013 annual report.)
Stock market declines ~13% from 24.86 to 21.71. Portfolio yield on original value: 3.4%
$1 million –> $900k
What we did:
– Winnie and I spent ~8 months traveling all over Mexico, 2 months in Guatemala, and 2 months in the US.
– We studied Spanish, ate insane amounts of good food, pickled our livers, and shed 16 years’ worth of work-related emotional baggage.
What we would do differently:
– Nothing. It was perfect.
Year 2, 2014 / 1930, Actual spending = $59,086 (See the full 2014 annual report.)
Stock market declines ~26% from 21.71 to 15.98. Portfolio yield on original value: 4.1%
$1 million –> $700k
What we did:
– Spent some time in Mexico and Cuba.
– Settled into Taiwan to do IVF / make a baby.
– Post-pregnancy we moved into a more expensive apartment, bought a couch, a flute, and 2 iPhones.
– I biked around the island and studied Chinese.
What we would do differently:
– Much the same, but wouldn’t have moved into the new apartment. I probably would have skipped the iPhone “upgrade.”
Reduced spending = $57,886 (<$2k savings)
Year 3, 2015 / 1931, Actual spending = $56,900 (See the full 2015 annual report.)
Stock market declines ~48% from 15.98 to 8.3. Portfolio yield on original value: 4.7%
$1 million –> $450k
What we did:
– Jr was born in April.
– We visited Japan in September and settled into Chiang Mai, Thailand in November. (Our apartment cost $375/month.)
– I attended a Chautauqua in October.
What we would do differently:
– OMG, I’m an unemployed deadbeat father and our portfolio has dropped by more than half! Shit! Time to panic!
– At least investment income exceeds our cost of living, and some of our recent expenses are 1-time events (IVF, childbirth.)
– Geographic arbitrage is already underway with the move to Thailand… let’s see how this plays out.
– We probably would have skipped the Japan trip and gone directly to Thailand.
Reduced spending = $53,300 (~$4k savings)
Year 4, 2016 / 1932, Actual spending = $72,002 (See the full 2016 annual report.)
Stock market declines another 15% from 8.3 to 7.09. Portfolio yield on original value: 4.8%
$1 million –> $460k
What we did:
– Traveled extensively. 1 month each in Thailand, Malaysia (& Singapore), Taiwan, 4 months in Western Europe (10 countries), 1 month in the US, back to Taiwan.
– Our Singapore hotel was the most expensive I’ve ever paid for (Marina Bay Sands with a rooftop pool.)
What we would have done differently:
– Instead of going to Europe we would have stayed in SE Asia or flown to Mexico / South America.
– Interestingly enough, we had originally reserved the same house we rented in San Miguel de Allende, Mexico for 6 months, but it fell through at the last minute and we decided to go to Europe instead.
Reduced spending = $42,402 ($30k savings, investment income still exceeds expenses)
Year 5, 2017 / 1933, Actual spending = $93,648 (See the full 2017 annual report.)
Stock market rose ~50% from 7.09 to 10.54. Portfolio yield on original value: 3.1%
$1 million –> $725k
What we did:
– 4 months in Europe (12 countries?), 1 month in the US, Alaska cruise, 6 weeks in Japan, back to Taiwan.
– Signed an 18-month lease on a 3 bedroom apartment in Taipei, doubling our rent.
– At his request, Jr starts full-time Montessori preschool at $1,000/month.
– I bought a sweet road bike.
What we would have done differently:
– Lots of things. Investment income dropped ~50% this year due to massive dividend cuts.
– Instead of moving around Europe and Japan, we likely would do SE Asia or South America, whichever we didn’t do the previous year. And traveled more slowly; speed = cost.
– Even though an Alaska cruise was Grandma’s dream vacation (our treat), I’m sure she would have been just as happy if we visited at her home for a couple of weeks.
– Assuming we settled back into Taiwan, we would have focused on the Xindian area (south Taipei) vs central Taipei. Several friends live there, and housing and schooling cost half as much. Honestly, we probably should have started there anyway, but now we are accustomed to life in the city center.
– I would have been happy with a much cheaper bike.
Reduced spending: $51,898 ($42k savings.)
Year 6, 2018 / 1934, Actual spending = ~$120,000
Stock market was down ~12% from 10.54 to 9.26. Portfolio yield on original value: 2.7%
$1 million –> $635k
What we did:
– Tried to have another baby (IVF.)
– When it didn’t work we went to Europe (9 countries) and the US (first time paying for housing in 2 places at once.)
– Spent Chinese New Year at beach resort in Vietnam.
– We stopped tracking every penny we spent. This easily increases our food spending by 10%+.
What we would have done differently:
– The main reason we came back to Taiwan at the end of the previous year was to try for another baby. Perhaps a cold thought, but if we weren’t experiencing irrational exuberance we probably wouldn’t have tried for #2.
– wouldn’t have purchased the ultra-lux handbag
– In all likelihood, we would have settled comfortably back into Mexico.
Reduced spending: $47,200 ($42k savings.)
Year 7, 2019 / 1935, Actual spending (estimated) = $100,000
Stock market rose ~50% from 9.26 to 13.76. Portfolio yield on original value: 2.7%
$1 million –> $875k
What we did:
– Hanging out in Taiwan, lots of biking, W is painting, Jr is in school, Chinese New Year in Thailand, summer in Vietnam and Bali
What we would have done differently:
– Same, but in Mexico with summer/holidays in coastal Columbia, Caribbean, etc…
Reduced spending: $48,700 ($46k savings)
Year 8, 2020 / 1936, Actual spending (projected) = $100,000
Stock market rose ~28%. Portfolio value is now greater than starting value!
$1 million –> $1.075 million!
Assuming we started spending an inflation-adjusted 4% of the original starting value from this point forward, after 30 years we would still have more money than when we started, albeit with continued large fluctuations.
Self Reflection
So… this roller coaster would have kinda sucked. (Albeit not too different from the worst time to retire ever.)
The good news is dividend and interest income exceeded expenses during each of these 7 years. Even so, it would have been emotionally challenging to stay the course with a substantial portfolio decline. As a new father, I probably would have looked at employment options (assuming there were jobs available.)
The bad news is there was very little we could do about it… a 75% stock market decline isn’t something that is overcome via reduced spending. Geographic arbitrage did help with the recovery, however. The conscious decision to spend well under 4% during the early years to reduce sequence of return risk is a good one.
Did we just get lucky, then?
Not really. The disaster scenario is rare.
The data for portfolio value after 7 years for every possible start date is interesting.
The 2013 retirement cohort is slightly above average but within 0.5 standard deviations.
Starting point: $1 million 90/10 portfolio with 4% withdrawal rate
Average after 7 years: $1.3 million
Median: $1.2 million
Portfolio has grown: 65% of cases
Portfolio has grown 50%+: 29% of cases
Summary
The doomsday scenario of portfolio collapse is scary, but rare and recoverable. Continuing to intentionally LBYM (live beneath your means) in the early years of retirement is a good tool to reduce the risk.
We were fortunate to see our portfolio grow in the first 7 years of retirement. This is the historical average.
Were the circumstances different, we would have made significantly different choices in terms of travel destinations, housing, eating habits, and even family size. Instead of increasing our expenditures with a growing portfolio, we would have limited spending to within 4% of the original portfolio value… life would still be great, just a little different. The portfolio would eventually recover, and then some. It looks like the 4% rule is quite robust.
Maybe tomorrow the stock market and global economy will collapse. If so, we know what to do.
This was the article I was hoping you would put together for awhile now. You one pays attention to Peter Schiff (predicted the house crash and predicts dollar crash) one does then need to at least plan for a potential massive hit. There are those who did not have parents that lived through the depression and WWII and don’t realize that shit does at times hit the fan..
Always a great perspective and analysis Jeremy ! Great articles as I get closer to joining the FIRE group. Thanks
I was kind of surprised, considering the overall market lost 75% I expected your 1 mil. portfolio to drop to maybe $300 thousand, but it looks like the lowest it got was $450 thousand. I guess those 10% bonds you carry really helped out in this scenario. Another well thought out article for the naysayers.
The stock market price dropped 75% but that period also had 25% deflation, so real portfolio value was about half (more or less.)
I think that adjustment for inflation is also killing my ability to look at this article.
It’s a weird thing… we experience stock market price changes in absolute terms. People just look at how many points the S&P500 has.
I have no idea what inflation is without looking it up. The Internet says prices have risen 11% in the 7 years since I quit work. I don’t think that matches my experience, but who knows.
I imagine if I was really going through the 1929 crash, I would feel that the portfolio dropped more like 75% than 50%.
I made a version of the chart with nominal prices.
Initial $1M, yearly 40k payments:
https://imgur.com/a/ByTyEuW
0% rate of return: exhausted after 25 years.
Good charts. It seems like if you can last 5 years, you’ll be able to weather pretty much any scenario. So it’s good to be conservative for the first 5 years. Once you’re over that phase, go crazy.
You’re right about reducing spending. We’re already at a pretty low level. We won’t be able to reduce much unless we change our lifestyle drastically.
Anyway, congratulations on 7 years. You’re over the hump.
But I think the “hump” would differ according to how long your retirement horizon is. The convention re: sequence of returns is, if you make it through the first 10 years smoothly on a 30 year retirement horizon, you’re golden. But that really seems to be saying you need to make it through the first 1/3 of retirement unscathed…not specifically 10 years. If you retire super early like GCC or RB40, then your retirement horizon will be significantly longer – maybe 50 years or more. In that case, the “hump” will surely be commensurately longer. Wouldn’t it?
It just depends on how long it takes to get to an actual withdrawal rate that will survive anything. That could be one year, it could be 30. In theory with savings of 25x annual spending, just putting that in TIPS it will last 25 years.
Listen to the mad FIentist’s interview of Kitces. He explains that the SRR is ten years no matter how long your retirement is.
Thanks for reflecting on this Jeremy. It’s actually quite good to see that even in a doomsday scenario of portfolio collapse you can manage to still have a pretty good life can’t you? We’ve been to Mexico for 4 months last year and can relate about the quality of life there. Another great example of the freedom you can get if you can apply geo-arbitrage to your life. We have been stow traveling full time for almost 18 months now and are a few weeks away for spending about 4-6 months in South East Asia. We can’t wait to see how far our dollar will get us :-)
It’s even better now. When we were in Mexico 1 USD bought you 13 pesos. Now it is 20.
So the market drops 75%, and then rises 200%, and your portfolio has only gained what 30% ? Not a great return in my books. If you had balanced your portfolio better, there would be cash to spend at the bottom and harvest some of those gains.
how would I balance my portfolio better?
I’m assuming “better” to mean safer or more conservative. If that’s a false assumption, you can throw my comment in the trash.
This is a double edged sword. The safer you play it, the more you give up in returns as the market goes up. And historically it averages an upward trend. If it were balanced more conservatively and he ran this same scenario starting at the bottom of a crash year, his overall return would be much worse. Risk vs reward.
Thanks for publishing this. I’ve been thinking about these various scenarios a lot lately, namely trying to recreate the math behind cFiresim… portfolio growth, yield, inflation, etc. and how each change year over year impacts portfolio balance. This is mainly for my own knowledge so I have more insight into the various scenarios.
Anyway, it looks like you are doing the same math here. Would you mind sharing the detail behind one year’s calculation? For example, in year one you list spending, market decline, and portfolio yield. How do these all come together to result in a $900K portfolio? I assume it’s pretty simple but seeing the equation would help. Thank you in advance!
See my reply to Rich below.
The bond math is more complicated as you need to also determine how bond prices are impacted by changing interest rates.
The forumula is discussed in a post on the bogleheads forum.
Thanks, this is very helpful. Just curious, is it generally accepted that the cFiresim math does all this more or less correctly? I wasn’t following closely when it came out. Was there validation done on the assumptions it makes?
Dunno.
I believe it to be correct based on building my own solution that gets the same output.
Pardon me, I don’t quite follow the calculations:
Year 1, 2013 / 1929, Actual spending = $38,966 (See the full 2013 annual report.)
Stock market declines ~13% from 24.86 to 21.71. Portfolio yield on original value: 3.4%
$1 million –> $900k
Year 2, 2014 / 1930, Actual spending = $59,086 (See the full 2014 annual report.)
Stock market declines ~26% from 21.71 to 15.98. Portfolio yield on original value: 4.1%
$1 million –> $700k
Year 3, 2015 / 1931, Actual spending = $56,900 (See the full 2015 annual report.)
Stock market declines ~48% from 15.98 to 8.3. Portfolio yield on original value: 4.7%
$1 million –> $450k
Year 1: $1,000,000 start + $34,000 dividends – $38,966 expenses – $126,710 stock market loss = $868,324
Year 2: $868,324 start + $35,601 dividends – $59,086 expenses – $229,180 stock market loss = $615,659
Year 3: $615,659 start + $28.936 dividends – $56,900 expenses – $295,886 stock market loss = $291,809
Where am I going wrong according to your calculations? Thanks!
I have the same question. Would love to see the detailed math behind these calculations.
There are maybe 4 things going on here.
1) The portfolio is 10% bonds
2) The numbers in year 2+ are adjusted for inflation (negative inflation in this case)
3) Numbers are loosely rounded
4) I listed our actual expenses but normalized the other numbers to a $1 million starting value, so it isn’t possible to recreate the numbers perfectly.
The Year 1 calculation with 4% withdrawal would be:
$1 million – $40k expenses = $960k; 90% stock = $864k, 10% bonds = $96k
Dividends = $29,888.97; Interest = $3456; Yield = (29888.97+3456)/1e6 = 3.33%
Stock decline = -109,477.07; Bond price change = $2,286.12
So, $1,000,000 start + $29.9k divs + $3,456 interest – 40k expenses – $109.5k stock market loss + $2,286 bond market gain = $886k (rounded to $900k)
I did this all in Excel, but you could get all of this from cFIREsim for a single year simulation (year 1929) with fees/drag set to zero (my Excel math matches cFIREsim perfectly.)
Here is the actual output file for that simulation.
In the even of a collapse such as you laid out the wife and I would probably rein in travel somewhat, our largest discretionary expense. We would still travel for months on end using our timeshare points, but when we do that we just use them as a second home, so costs don’t rise much over staying put. Things like cruises and the like would be dropped. No debts including a mortgage, and living in one of the lowest COL areas in the US makes our fixed expenses lower than most people. We have two SS checks coming in, and the wife has a pension, so in a pinch we can get by fairly easily compared to the majority of Americans. Best of luck to you and your readers in the coming year, which I believe will still show gains for the market. So let it be written, so let it be done!
Can’t get more solid than that
When it looked like going to hell in late 1980, with gold price plunging and USA interest rates sky-rocketing, I put all wealth in a Merrill Lynch cash management account paying ~18% compound monthly and set off around the world for 2.5 years – a year of which by camper van in the back blocks through the ‘hell’ of a neigh-on depressed USA. On returning to Aus, employment had improved so I worked.
Which hell? Economic hell like USA 1981, 1990, 2001, 2008? Reducing interest rates and no economic hell in Aus since 1990 has resulted in households accumulating large debt (and assets) which are an economic vulnerability. I’ve ‘prepared’ with access to gold which could cover 2-4 years of normal to emergency expenses in a economic hell – useless against an Apocalypse. Cash in the virtual saddle bags to ambush opportunity. In the meantime I don’t feel pressured to spend.
We went to the UK post Brexit vote when the GBP dropped 10-15%. Now seems like a good time to visit Argentina
Argentina better by far by drone (unless you take one): https://tinyurl.com/y28ws5py
I’d cut back on the fine wine. Otherwise, my goal is to spend more on wine than property taxes…
#goals
I’ll be honest. The year 3 drop when the portfolio is reduced by more than half would have truly freaked me out. In hindsight it is always easier to look back. I remember buying Citi during the Great Financial crisis as it kept dropping and at some point I wondered if I was being crazy with the risk taken.
I almost bought a bank ETF that was yielding 10%+ in 2008 but chickened out. Oops, missed out big time.
Thanks for this analysis Jeremy.
And especially, thanks for the perspective.
Thank you for doing this detail analysis. Very helpful!
You forgot to mention one other obvious solution to a tanking market and panic-level anxiety over running out of money- get a job! It wouldn’t even have to be going back to your old job. You could teach English in Barcelona for 15-25€/hr depending on your clientele. You could teach skiing in California for the same rate (granted you’d be in expensive California so you might want a ski bum roommate).
People who are savvy enough to retire years/decades ahead of when society says retirement is “acceptable” are, no doubt, going to approach a tanking market in an unconventional way as well. Resilience and not giving a rat’s ass about convention can overcome most any stock market crash.
I loved this article as it clearly shows that even if you had retired into a crashing market you still wouldn’t have to get a job (gasp!). That’s apparently the worst thing an early retiree can imagine!
I probably would have looked for work after Jr was born.
Another beautiful bit of writing and analysis. Thanks very much, Jeremy.
It’s also interesting to think about taxes during a worst case scenario like you’ve illustrated. No capital gains means no capital gains taxes during (hopefully minimal) sales of stock index funds for cash to cover living expenses.
While we obviously couldn’t harvest capital gains, Traditional-to-Roth IRA conversions would still be desirable during that interval. And you’d probably enjoy higher ACA subsidies–MAGI would be easily less than the ACA cliff. This could help reduce expenses during the “dark days.” Cold comfort, I know, but better than nothing.
Keep up the great writing, and thanks again.
yessir, it is always interesting to think about taxes :P
A great benefit of capital gain harvesting is you always have some amount of stock at full basis. When the market drops you can then harvest some losses, and use $3k of them to offset other non-capital gain income… thereby increasing your annual Roth conversion by $3k or lowering MAGI.