One of the central tenets of the FIRE community is the 4% rule, aka the Trinity Study. For those of you who are new to FIRE the 4% rule as understood by at least some people in the community is that you can safely draw down 4% of the amount invested at the start of your early retirement and never run out of money. This is based on a study done at Trinity University in the US looking at historical data for a 50/50 portfolio of stocks/bonds running over 30 year rolling time periods with the last period ending in 1995. What doesn’t get mentioned, particularly here in Australia, is that the study is based on US data and that in at least some of the 30 year time periods the money ran out, and also that it’s only for that 30 year period rather than in perpetuity. So that nice safe 4% withdrawal rate isn’t actually based on Australian data and ran out of money in some scenarios. On top of that if there was still $1 left at the end of the 30 years it is classed as being a success even though come year 31 you’re going to be in a world of hurt? Are you starting to worry a little bit about retiring early based on the 4% rule yet?
Researchers at Griffith University in Australia produced this paper (https://www.griffith.edu.au/__data/assets/pdf_file/0024/205683/FPRJ-V1-ISS1-pp22-32-safe-withdrawal-rates.pdf) for the safe withdrawal rates in a number of different countries including Australia. Based on the historical data for a 50/50 portfolio of stocks/bonds the safe withdrawal rate in Australia for a 30 year period was only 2.96% rather than 4%.
If you used a 4% withdrawal rate then over a 30 year period 18% of the time all of the money was exhausted, and over a 40 year period 42% of the time all of the money was gone. Even over a shorter 20 year period 2% of the time all of the money ran out. And even if the money didn’t run out it’s not covering you all the way through life, only for that 30 or 40 year period. As I said above if you have $1 left at the end of 30 years then the study says it’s a success even though in year 31 it’s going to be pretty tough going. On top of that this is all based on historical data and there is absolutely no guarantee that the future will look like the past. That 4% rule sure looks a bit shaky!
To add fuel to the FIRE (pun intended) in fact even at a 3% withdrawal rate over a 40 year period the money was all used up 7% of the time, and 1% of the time over a 30 year period. Which sounds like reasonable odds and maybe are, but it’s not going to be much fun if you’re the one who gets the unlucky portfolio. By the way all of this is assuming a steady 4% drawdown on the initial amount invested, and doesn’t take into account any lump sum withdrawals for things like a car that needs replacing, an expensive overseas holiday. So even if you did manage to have one of the 30 year periods that worked out just fine, it’s not accounting for any lump sum withdrawals which you’re likely to have to make over time.
Now there’s all sorts of caveats to the above, some of which support the FIRE 4% interpretation and some of which don’t. Even the most aggressive FIRE practitioners are generally only looking at retiring in their mid to late 30s with most people looking at their early 40s or beyond. Australia’s age pension kicks in at age 67 and provides approximately $35,000 a year for a couple who own their own home and have less than roughly $350,000 in other assets. So if you’re retiring at 37 and make it through your 30 year period to age 67 you’re probably going to be ok although you’re unlikely to be living it large on that amount of annual income.
The typical investment portfolio is also unlikely to be split 50/50 between shares and bonds, most FIRE practitioners advocate having more like an 80/20 share/bond split if not all shares. That in itself has some issues depending on what you are invested in and how much of an income stream those shares are producing. If you’re mostly in shares that provide capital growth rather than income then you’re going to need to be selling off shares to fund that 4% withdrawal rate. If you’re in shares that mostly produce income through dividends then you’re probably in a better place, except what happens if a recession hits and dividends get cut? There is usually less of a reduction in the dividend than there is in the share price, but looking at this dividend chart (http://www.sharedividends.com.au/STW) for STW which tracks the ASX200 dividends seemed to get cut enormously in 2009.
So if you’d retired in 2007 0r 2008 and been relying on that income stream from dividends and it was cut by 60%, then you’re selling shares to fund your living costs and doing so at what turned out to be the bottom of the market. The STW share price fell about 50% from peak to trough so you’d have been combining a huge cut in dividend income as well as selling down your shares at half their previous value which has to hurt psychologically. In which case that 4% rule likely did not work out too well for you!
Unfortunately I don’t have data for Australia but this article has some of the data for the US which shows that for a 75/25 shares/bond split you’re looking at a non zero failure rate after 30 years if you’re drawing down 4% of the initial capital each year. https://www.forbes.com/sites/wadepfau/2018/01/16/the-trinity-study-and-portfolio-success-rates-updated-to-2018/#720dab236860 Interestingly the failure rate is actually lower after 40 years for investments with all shares or 75% shares than a 50/50 portfolio. As a side note it’s not an apples vs apples comparison between Australia and the US given the much higher dividend yield of about 4% here (closer to 6% with franking credits) vs around 2% in the US which means that you would almost certainly have to sell off shares in the US at some point vs conceivably not having to do so in Australia.
If you’re looking at using rental properties to fund your living expenses using the 4% rule, then that also has issues. You probably need to have paid them off in full otherwise your rental income is being used to pay the mortgage rather than providing income for you so forget having any mortgage left on them. On top of that if for whatever reason you don’t have a tenant in the property then you don’t have any income coming in at all but you still have expenses. This might not happen very often but it does happen from time to time.
And last but certainly not least, median rental yields across Australia are only 3.63% and that’s gross, in both the sense that it’s pretty low and that it is before any costs, taxes etc are taken out of that yield. (https://www.businessinsider.com.au/australian-property-rental-yields-2018-1)) So you’re not getting 4% income here, and it’s not as though you can just sell off part of the property to help fund your living expenses. Admittedly regional yields are higher at 4.94% compared to 3.32% in the capital cities but even then your cost may pretty quickly get your income level down below that “safe” 4% rate that you’re relying on.
The other usual caveats are that those who are already retired on FIRE would just get a job if they weren’t meeting their living expenses, which makes sense except that the time that is most likely to happen is in a recession, in which case jobs are going to be pretty hard to come by. Alternatively there’s the “I’ll just tighten the belt a bit” which is great except that there is only so much of this that you can do given that a lot of living costs are fixed or near enough to it (food, rent/mortgage, utility bills etc) and most FIRE practitioners are looking at fairly low incomes in their early retirement already so there isn’t much room to tighten that belt. See my post on fixed living costs for more on this (https://aussiehifire.com/2018/08/17/fixed-living-costs/).
Also last but not least, the 4% rule is the gross amount you’re taking out of the portfolio and doesn’t account for any taxes you might be paying. Now admittedly if you’re looking at using say a million dollar portfolio to fund $40,000 in living costs and it’s split between two people then you’re not going to be paying much if any tax at all. But if for whatever reason most of the portfolio is in one person’s name then you will be handing some money over to the tax man, which means you need to have even more money saved.
None of this is to say that the 4% rule doesn’t work at all. There are plenty of scenarios in which it works out absolutely fine and an early retiree lives off it very easily and in fact grows their money considerably over time which is obviously the ideal scenario and I very much hope that this is the way things go for FIRE participants. But hope is not a strategy, and the 4% rule is unfortunately not guaranteed to work.
Speaking for myself I still plan on using something similar to the 4% rule and living off dividends rather than drawing down on my capital as much as possible. But instead of aiming for the bare minimum and hoping for the best I plan on having plenty of safety margin so that if something does go wrong there is plenty of room to tighten the belt rather than having to sell off shares or go and get another job at the worst possible times for both.
What do you think about the 4% rule? If you liked this post and would like to read more like it then please subscribe!
Interesting post and perspective. What are your thoughts on holding cash reserves for say a year or two of expenses to help make it through a downturn during FIRE?
I wonder how the success rates would change with various cash holdings. On one hand you’d have reduced returns, however you wouldn’t need to draw down on capital in the event of reduced dividend yield.
Cheers,
MC
Hi MC
Good question! As you’ve said there is definitely a tradeoff there if you hold a year or two worth of cash. You’re almost certainly going to have lower returns over the longer term by having that cash, but you don’t need to sell shares if dividends get cut. Personally I prefer to limit my downside and if that means lower returns over the long term then so be it so I plan to opt for having a couple of years worth of my fixed living costs in cash. My fixed living costs are a lot less than what I plan on having income wise anyway so it won’t change my overall asset allocation much.
Depending on what your asset allocation is though you may already have that much cash anyway. The study that the original 4% rule is from assumed a 50/50 equity/bond split and given the very minimal spread between cash rates and bond yields at the moment you might as well park at least some of that money in cash so that it is easily accessible if you need it and there isn’t any real change in your asset allocation. The same applies for pretty much any asset allocation apart from 100% equities. If you want to get technical by having money in cash/term deposits rather than bonds then you miss out on any positive effects from duration if yields change but you also miss out on the negative ones. Given where bonds yields are at the moment I think the risk is to the downside rather than the upside anyway but who knows.
If you’re going for 100% equities and using the 4% rule then putting two years worth of your overall money in cash only drops your equity allocation to 92%, so to get a rough estimate of the drop in performance multiply whatever you were using for your rate of return by 0.92 to get your new rate or return. This doesn’t factor in whatever return you get on your cash though which probably bumps it back up slightly anyway. And to be honest if a pretty small difference in assumed rates of return is going to skupper your FIRE plans, then maybe they aren’t that robust to begin with? This article gives the changes in success rates for the US if you go from 100% equities to 75%, depending on the time frame having that extra money in bonds seems to boost the success rate by about 3 to 4%. Obviously that is US conditions though rather than Australia.
So I guess long story short I don’t have the answer but it seems likely to me intuitively that you would boost your success rates by having at least some of your assets in cash/fixed income.
Check out https://earlyretirementnow.com/2017/03/29/the-ultimate-guide-to-safe-withdrawal-rates-part-12-cash-cushion/ re cash
If you look at the safe withdrawal rates series by early retirement now (earlyretirementnow.com) he shows that cash is basically always bad.
He does talk about a glide path using some bonds to smooth out the start of retirement and decrease sequence of return risk.
Anyway have a look at the blog, lots of great info there with tons of mathematics to show his points.
Hi Andrew, I’m glad you enjoyed the post and thanks for for the link, I do recall looking at it a while back. As per my post I agree that dividends can be cut, although they do tend to be a lot steadier over time. I’m still trying to find out why STW had such a dramatic drop in dividends given there didn’t seem to be anywhere near that large a drop in the dividends from the underlying shares, I’m guessing that they must have been including some capital gains in the previous payouts. I called up State Street and the registry to check if perhaps there had been a share split but apparently this wasn’t the case so the drop is real even if I’m not sure of the reason for it. In any case while I agree with the maths from your link, the obvious caveat is that he is looking at the US which has different conditions to Australia so what works there won’t necessarily work here but that’s not to say it wouldn’t apply to Australia anyway.
I think you’ve got to look at what the market conditions are at various stages and decide what you want to use to ride out the rough patches. As I mentioned in another comment all bonds aren’t equal so you really want to think about what you own and how it will behave. Treasuries are likely to increase in price in a crisis with a flight to safety, High Yield will go the wrong way, and IG can go either way but probably down. It would also be interesting to look at the historical premiums of bond yields over cash or term deposit rates. As things currently stand I can get a better return in a TD or HISA than by buying government bonds and have very little duration risk, although if rates drop the government bonds might do better thanks to prices increasing. So there is that to consider as well.
As I said I plan on keeping some cash in the portfolio and if necessary topping it up from time to time. If I assume very little return from that and plan on living purely off dividends then I should have a bit of a safety margin if dividends do get cut, although I would probably want to have a look at what amount of the dividends coming from my portfolio (assuming I’m in index ETFs) reflect dividends from the underlying holdings and what amount is coming from realised capital gains and is likely to very quickly dry up if the market crashes. I’ve still got a while yet to think about things in any case!
as you have mentioned in comments. In a split portfolio 50/50 80/20 share/bonds- as your share price dropped you would actually sell your bond side to try to even up the portfolio percentages.
It’s also important to remember that if you are using actual bonds or a bond fund there’s no guarantee that in a crisis that portion of your money won’t be losing money as well. As a rule of thumb sovereign debt tends to increase in value in a crisis because there is a flight to safety, high yield debt behaves more like equities and decreases in value. Investment grade debt can and does go either way depending on how the credit of the company holds up. So ideally you might have some of your fixed income money in actual cash, especially given the extremely flat yield curve at the moment. You can then use that cash to tide you over until markets recover.
Really nice post, thank you for the linking the paper too.
Pingback: How would the proposed changes to imputations credits affect FIRE? | Aussie HIFIRE
I enjoyed reading this blog post. I don’t think this area is getting enough attention due to it being such a long time since investors have run into any problems. It is easy to theoretically say things such as we can ride out the bear market, tighten our belts, go back into the workforce etc. But the only people that can perhaps truly be confident of this are those that say achieved F.I.R.E. pre GFC and did so. For the rest of us we are simply applying some neat theories with the hindsight of the past, and using an analytical way to speculate what we may do in the future under very emotional circumstances.
I also think the 4% rule is on the high side, and we get overly comfortable with this from cherry picking the U.S. & Australia equity market long term returns, which happen to be virtually the best of the world in the last century.
One question I had was if you believe the drop in STW dividend was the correct example implying 60% cuts to Australian dividends. I don’t have the data on hand but thought the drop in total dividends was no where near as sharp as the share price falls? Maybe only circa 20%?
Hope you keep the blog going, good work.
Hi Steve, thanks for reading! As you say investors haven’t run into much trouble for a long while now, and I think a lot of the FIRE community are young enough that they weren’t even looking at what was happening to markets 10 years ago. Speaking as someone who was working on a trading floor at the time, it was scary as hell and as much as people think they will just ride it out it will be incredibly difficult mentally to do so. Maybe a subject for a future post!
I talk a little about the actual drop in dividends from companies in another post here https://aussiehifire.com/2018/08/20/sequencing-risk-dividends-and-retiring-at-the-end-of-a-bull-market/ but I had to rely on eyeballing the graph from the RBA paper rather than having the actual data to play with. It looks roughly like a 10% drop to me, no more than 20% in any case. I’m assuming that a lot of the increase in the STW dividend prior to the GFC was capital gains being distributed, share buybacks, special dividends etc and then when the GFC hit all of that went away.
Pingback: Why I don’t automate my finances and pay myself first | Aussie HIFIRE