Making a bigger safety net – how much longer does it take to get to a 3% withdrawal rate

There’s been a lot of work done recently on sequencing risk and safe withdrawal rates for Australian investors.  Dan over at Ordinary Dollar has done some fantastic work with his series and I’d like to think that I’ve contributed something to the FIRE community with my work as well. 

One thing that has come out of this is that the magical 4% withdrawal rate using only Australian investments has not historically been 100% safe.  I talked about it in this post here and Dan covered it in this post.  What we both find is that if you are only looking at Australian investments then you need to be more conservative with your withdrawal rate.

Obviously the safe withdrawal rate varies somewhat depending on the actual mix of investments but as a generalisation 3% seems to work pretty well across asset allocations where most of the money is in Australian shares.  Which happily enough matches up with what some academic researchers found as well.

So if you want to reduce the odds of running out of money in retirement then you need to be using a more conservative withdrawal rate.  This means that you can have the same amount of money and retire at the same time as you would have otherwise but on a lower income, or retire later but on the same income you would have wanted at a 4% withdrawal rate but using a 3% withdrawal rate instead.  And obviously you can do a combination of these two things as well. 

Assuming that people want to be living a pretty comfortable life in retirement and don’t want to be starting out with less room to tighten their belt then they will likely be looking to save up more money before declaring FIRE. 

How does it work

If you want to use a 3% withdrawal rate instead of a 4% rate then the simple math is that you need an extra third of the amount of money you needed previously.  How much longer it will take to get there though is much more complex. 

The simplest and easiest way to try and figure this out is to put a bunch of assumptions into a compound interest calculator like this one at Moneysmart and let it do the math for you.  Which as I said is simple and easy which is always nice, but more importantly it’s almost certainly wrong.

The reason for this is that as discussed in this article returns are not smooth and instead vary over time.  Some years they’re up, some years they’re down, and as you get closer to your target it’s entirely possible that despite investing more money each year you might actually have less money at the end of that year.  This is due to the returns on your existing investments being more of a factor than whatever additional amount you’re contributing, as I talked about in this article.

So using historical stockmarket data, how much longer would it have actually taken to build up the additional funds required to use a 3% withdrawal rate rather than a 4% withdrawal rate for the same amount of income?

Obviously there are a few variables that need to be considered here, primarily what the actual additional savings each year will be as well as what the amount of income desired in retirement is.  Assuming the same variables for both a 3% withdrawal rate and a 4% withdrawal rate we can then see just how much longer it would have taken to get there.

I’ve used scenarios with $30k, $40k and $50k per annum as the targets for hitting FIRE to represent roughly speaking my versions of Lean, Regular and HIFIRE respectively (assuming a paid off home).  You may have different version of the dollar amounts needed and that’s fine, these are for illustrative purposes. The table above shows what amounts are needed to be able to withdraw the respective amounts at each withdrawal rate.  I then also used annual savings rates of $30k for the $30k and $40k withdrawal rates and a $40k annual savings rate for the $50k withdrawal rate to calculate how long it would take to reach FIRE. 

What do the results show?

The table below shows the number of years taken to hit targets of $750k and $1m assuming an annual contribution rate of $30,000.  These targets would then generate an income of $30,000pa (lean FIRE) at a withdrawal rate of 4% and 3%. 

We can see that for a lot of the starting years the gap between how long it takes to hit FIRE is actually very small even though another third more money is required.  Just under half the time the gap is a year or less, about two thirds of the time it is under two years. 

The scenarios targeting $1m or $1.33m to generate a $40,000 income at withdrawal rates of 4% and 3% also used a savings rate of $30,000 but as would be expected took a little bit longer to get there given the savings rate was a lower percentage of the total. 

Again though for a lot of the starting periods there isn’t much of a difference in how long it took to hit FIRE.  Similarly to the $30k withdrawal scenario just under half the time the gap was a year or less.

Lastly the target of $50,000 income in retirement required portfolios of $1.25m and $1.66m at 4% and 3% withdrawal rates.  I assumed a $40k contribution rate here rather than the $30k on the other scenarios on the assumption that anyone who wanted this higher income was probably already on a high income while they were working and could probably save a bit more because of this.

This time in just over 40% of the starting periods it only took a year or less to hit the higher target, just under 60% made it within two years.

The graph below shows the actual gap between various scenarios over time and as you can see much of the time it is quite low, in fact the average is under 2.5 years although there are a number of much higher gaps so you would have spent a lot longer waiting to hit the higher target and have a safer withdrawal rate.

Why is the gap so variable?

So why is it that despite requiring a third more money to hit FIRE about half the time it took a year or less to reach the higher goal?  There are two main factors at play here.  The first of these is that once you get to a certain size in your portfolio, the returns on the portfolio itself swamp the contributions you are making so they become less relevant.

The second and more important factor is that those returns tend to be lumpy.  Over the time periods that we’re running the calculations on here the average return after inflation is just under 10%.  But the number of years that the actual returns post inflation are lower than a 10% swing either way is actually less than a quarter of the time. 

When the stockmarket is up, it tends to be up big which is great fun!  And on the flip side when it’s down, it tends to be down big as well which is obviously not so enjoyable particularly if you’ve already got a lot of money invested.

What this means for us is that if you put a couple of positive years together then anything that’s remotely close to your FIRE target at a 4% withdrawal rate will hit it, and there is also a reasonable chance (about 45% on the scenarios I ran) that the 3% withdrawal rate will do so as well.  However if you do just miss out and straight away hit a big down year, then that may put you back several years.  This is sequencing risk at work. 

There are plenty of examples where it took 5, 6 or as many as 9 extra years to hit the amount required for a 3% withdrawal rate rather than 4%.  When this happened it took far longer than an extra third of the time to hit FIRE for a 3% withdrawal rate, in the most extreme case getting that extra third more money required actually took another 60% of the time to get to the 4% withdrawal target amount (15 years to hit 4% target, another 9 years to hit 3% target so 24 years in total). 

This particular case was due to the lower amount needed for FIRE being reached just before the GFC with the higher one just missing out and then getting devastated in the GFC.  This type of event was generally the case where any two scenarios had a big gap between when they hit FIRE, there was a big downturn in the market which meant that the second portfolio took a number of years to recover.

The other side of this would be that someone who had just hit FIRE at their 4% withdrawal rate pre GFC immediately experienced a big downturn in the value of their portfolio AND they had presumably quit their job already so it would have been nerve wracking times.

I’m not sure which scenario I prefer, being retired and no longer having to work but immediately worrying about whether I will have enough money in retirement, or still having a reliable source of income through employment but having to work for another 8 or 9 years? 

And of course in the first instance if I’d been able to look into the future and see this article I’d have known that returns in the first five years of retirement are a huge predictor of whether you will run out of money or not, I might have been more worried still!

Clustering – why everyone hits FIRE at the same time

Another consequence of the lumpy returns is that we see clustering of when scenarios actually hit FIRE.  I talked about this phenomenon and as you would expect we see the same thing again for these scenarios as shown in the chart below.

Lots of scenarios hit their targets in 1979 and 1980, then a bunch more in 1985 and 1986, and then few more in 1989 to round out the 80s.  In the 90s there were a lot that got there in 1992 and 1993, a gap in 1994 when the market was down enough that it overcame any additional investments and  then it was pretty smooth sailing the rest of the way through the 90s and 2000s.  Then the GFC came along after which no scenarios hit FIRE for quite some time because they had to make up their portfolios being cut in half back in 2008.

What can we learn from this?

Working and saving for another year or two might make a big difference to your margin of safety.  Just under 60% of the time working for another year or two takes you from having enough at a withdrawal rate of 4% to having enough at a 3% withdrawal rate whilst taking out the same amount of money.  That’s a huge extra margin of safety for sticking with work for just another year or two.

The flip side of this is that if you’re hanging out for that extra margin of safety it might take you a lot longer to get there, potentially up to several years depending on the scenario, although it almost certainly means that you’ve dodged retiring just before a big downturn in the market, so that’s presumably pretty comforting.

The ultimate lesson though is that as I’ve said in previous posts, sequence of returns is an absolutely huge driver of when we can retire and how successful that retirement will be.  How much you’re investing each year obviously makes some difference to how much money you have, but once your portfolio gets to a reasonable size the movements of the market will almost always have more of an impact whether it be positive or negative.  Our results go from being mostly within our control to being primarily driven by market returns.

Will this change what I do?

Despite knowing the futility of running the calculations given I have no idea what future returns will be and they almost certainly will not be the same as past ones, because I’m as curious as the next person I have of course run these historical scenarios on my own situation.  Interestingly enough in three quarters of the scenarios it took me a year or less to go from hitting my target at a 4% withdrawal rate to hitting it a 3% withdrawal rate. Yay for me! Well, me and the family I suppose but at the moment it’s just me working so let’s go with mostly me.

So it’s possible that I may work for another year post hitting my FIRE target at a 4% withdrawal rate.  This would help by having another years worth of contributions, but it’s also another year that I’m not drawing down on my portfolio so to some extent it’s a double whammy in my favour. 

On the other hand at that point the biggest determinant of what happens to my portfolio is likely to be what returns the market has rather than anything I do.  So it’s likely to come down to how I am feeling about work at that point in time, and there is absolutely zero point in locking myself into a decision about it now!

Are you aiming for a 3% withdrawal rate or 4%?  And has this post changed your mind on this?  If you enjoyed this post and would like to read more like it then please subscribe!