As I’ve talked about a number of times previously returns aren’t smooth and so we have to worry about sequencing risk ie a big fall in the value of our investments at the wrong time derailing our plans for FIRE. As much as it’s nice to think that we can control our journey to FIRE, the reality is that so much of the outcome is driven by the returns of our investments.
For the first 3 or 4 years these don’t matter so much, whatever we are putting into investments is usually far more than the effect of the returns of those investments. After those early years though the returns on our investment can get big enough to determine if the value of those investments went up or down and it makes more of a difference as we have more money invested and get closer to hitting our numbers. So what effect does that have on when we can retire?
Using historical data I wanted to have a look at this across a range of different inputs so I’ve gone with annual savings rates of $30k, $40k and $50k as well as targets of $1M, $1.25M and $1.5M for a total of six different scenarios. This is hopefully pretty much in line with what most of the FIRE community are aiming for so that it’s reasonably representative. Unfortunately, the results are not what we would like to see.
The scenarios I ran started in 1959 and assumed that whatever amount you had managed to save over the year you’d put into the stock market in one lump sum at the end of the year. I considered using a mix of shares and fixed income and trying to maintain some sort of reasonably constant asset allocation but decided not to for two main reasons.
Firstly, plenty of people in the FIRE community seem to be going for 100% shares or close to it so it wasn’t going to be as relevant. Second, and more importantly to me at least, to be honest it would have been a pain trying to write the spreadsheet for it and I have limited time for this sort of stuff. So here we are.
I also assumed a constant (inflation adjusted) amount of savings each year. Which yeah, let’s be honest it’s not likely to work out that way in the real world. It is however easier to calculate and given the savings amount is going to be different over time for everyone anyway as they get raises, change jobs, have kids and take leave, go part time etc I figured I might as well use a constant savings amount.
All that said, what were the actual results? The chart below shows the number of scenarios which hit their FIRE goal in each particular year.
I’ve started the results in 1970 but the scenarios themselves started in 1959 and as you can see with just two exceptions none of them hit FIRE until the end of 1979. It was actually taking up to 25 years depending on when you were starting to hit FIRE (I’ll be doing a post on how many years it would have taken historically in the near future).
The reason for this is the big down markets in 1973 and 1974 which would have more than halved the value of people’s investments at that point.After that there were a number of years of good returns but it still took 5 or 6 years to get back up to hitting FIRE targets, but because those years had been so good a number of scenarios would have all hit FIRE at once. albeit with different amounts at the end although all over their target.
The early 80s weren’t that great for hitting FIRE due to big downturns in the market in 1981 and 1982 which put off retirement for a few years, but there were huge clusters of hitting FIRE numbers at the end of 1985 and 1986 because of some stellar equity returns in those years.
Although everyone remembers Black Monday in 1987 (or has heard about it more likely, the FIRE community skews pretty young I think!) the actual year wasn’t that bad so there were still plenty of people hitting FIRE a couple of years later although none in 1987 itself.
The nineties are perhaps best remembered for grunge music from the likes of Pearl Jam and Nirvana which at least some people found a bit downbeat (I was more of a Goo Goo Dolls or Counting Crows fan myself), but it was a pretty good time for FIRE. Stellar returns in 1992 and 1993 helped out of course and there were a lot of people hitting FIRE then, but most other years people were hitting FIRE on all or most of the various scenarios.
The early 2000s were also good years for FIRE with a steady stream of people making it. And then along came the GFC in 2008. Having the value of your investments pretty much halve is not going to be the good news for anyone but it was pretty surprising to me that it put off FIRE for another 5 years minimum.
And it’s not as though afterwards we start seeing a bunch of people all hitting FIRE at once, because the returns in the teen years (when are we going to come up with a good name for these years?) have been fairly steady rather than big movements one way or the other.
The above chart shows the number of various scenarios hitting FIRE in any given year and as I’ve talked about above we see a lot of clustering around 1979 and 1980, then again in 1985 and 1986. In the nineties 1992 and 1993 were also good years with a lot of difference scenarios hitting their numbers and then fairly steady progression the rest of the time. The early to mid 2000s were fine to good, and then the FIRE went out completely (See what I did there? Can you tell I’m a dad with these jokes?) from 2008 through till 2013.
So keeping in mind that the scenarios all started in 1959, it obviously took a while for some scenarios to hit FIRE and there were obviously a lot of years or years in a row where nobody at all would have hit FIRE. The 70s were a brutal period, the 80s had some good years but also a lot of years where nobody hit FIRE, sometimes for a few years in a row depending on the scenario. The late 2000s were terrible as well with nobody hitting FIRE for 5 years straight from any scenario.
As I said at the start I’m using a 100% Australian equity portfolio here so obviously these scenarios don’t apply to everyone (or even maybe anyone) but it does show that it’s not going to be a nice predictable process of putting x amount in each year and hitting FIRE at precisely 13.37 years. And yes I do realise that if you were invested in a diversified portfolio or some other asset class then your results would be different.
The point is though that as much as we might like to think we are in control of our own destiny, when we hit FIRE is very much going to come down to the returns of whatever we’ve invested in. This is due to the returns on our investments likely making much more of a difference to the value of our investments than whatever we are adding to them. A big fall at the wrong time might derail our FIRE plans for several years.
Although historically I’ve leaned towards being mostly in equities, I’m now rethinking whether this is going to be the best approach and at some stage will need to run some numbers on this. Having a more diversified portfolio may mean taking longer to get to the destination (although it may also be quicker), but it’s also likely to reduce the chances of having it snatched away just a year or two out from hitting FIRE.
What are your thoughts? If you enjoyed this If you liked this post and would like to read more like it then please subscribe!
Excellent analysis HIFIRE! This is really timely as I review my own asset allocation in my arguably over-diversified portfolio. It’s very easy to imagine my FI date receding a few years in the current environment. Do you think that they’d be any value in a version that randomized the year return order, to see if the same finding would hold so strongly? Intuitively, you’d imagine it would, i.e. lots of FIREing just before the storm…
Hi Fi Explorer! You’re certainly got quite an interesting portfolio that’s for sure! What changes are you thinking about making to it? As I recall you’ve actually achieved your first goal on an income basis which is the more important aspect to be looking at in my opinion, although presumably some of that at least is capital gains being distributed to you rather than dividends etc generated from the underlying holdings.
To some extent these scenarios are already randomised because they start in different years with different savings rates as well as the actual returns being random in the sense that the market and it’s returns are indeed somewhat random. I did a top to bottom and vice versa version of this for this post but over a shorter time frame and it certainly makes a difference. Does that answer the question or have I misinterpreted it?
Thanks, yes, it was that sequencing point I was drawing out. I think you’ve answered it.
At the moment I’m thinking through changes to my portfolio, but one significant change is likely to be some more explicit risk-reward tradeoff around foreign equities exposure. But still deep in thinking and research, so there could be others.
One of the reason why dividend growth style of investing in LIC becomes so popular for it mainly focus on income, not just the ever changing ebs and flows of the shares market price. So if one is receiving dividends and thats enough for him to live on then it doesnt matter whether whether your in a bull or bear market as you dont have to sell shares, just live on dividends
I agree that dividends don’t fluctuate as much as actual share prices, but dividends can and do get cut, and it’s usually when the market is falling. Given that in theory at least share prices should reflect future expected earnings share price falls should be a trailing indicator of future expected earnings and dividend payouts. I talked a bit about dividends getting cut in this post a while back.
It’s worth looking at this analysis in light of your Sequencing Risk analysis. Let’s look at the 4% withdrawal figures for 100% stocks from the Sequencing Risk and remove any years when we don’t hit FIRE according to your analysis above.
1972/1973*: $373,782.70
1979/1980: $10,033,237.93
1980/1981: $5,585,398.26
1983/1984: $8,974,168.78
1985/1986: $6,835,194.40
1986/1987: $3,832,411.94
* the two years on each line are the years used in each blog post. e.g. FIRE at end of 1972 (this post) = Start retirement in 1973 (Sequencing Risk post).
The best performing years are gone, including all six years with final worth above $15M. But they are an illusion as we wouldn’t have hit FIRE then and so would not have started retirement.
Good to see that even when the best years are removed there is still money left after 30 years, even if it’s not as much on average as we might have thought.
Damn it, that was going to be another post! 😉 As the numbers show all of the people who declared FIRE would have made it through a subsequent 30 year (or longer) period. I don’t have the numbers in front of me but the 1973 FIREys would have been a pretty scary ride but anyone who did it would be well into their 70s at a minimum now I guess and getting pretty close to a full age pension so would be quite content I’d imagine. And in any of the other scenarios you would be a very happy little camper indeed!
Pingback: How long will it take to hit FIRE? | Aussie HIFIRE
Pingback: The shockingly complex math behind early retirement | Aussie HIFIRE
Pingback: Financial New Years Resolutions | Aussie HIFIRE
Pingback: Safe Withdrawal Rates for Aussies — Part 7: Sequence of Returns Risk – Ordinary Dollar