Today’s post is our fourth visit to the Science of Retirement Conference 2018, which was held in London last spring. This time we’ll be looking at Improving Retirement Income.
Improving Retirement Income
It’s been quite a few months since I wrote the three previous posts about the Science of Retirement Conference (SORC), which I attended last spring.
- In the meantime, we’ve covered ERN’s mammoth series of posts about Safe Withdrawal Rates.
I know a lot more about decumulation than I did when I attended SORC.
- So it will be interesting to revisit Abraham’s presentation, which at the time I found very informative.
The last presentation of the morning was by Abraham Okusanya, the organiser of the conference.
- We’ve met him several times before, as he’s one of the biggest names in decumulation on this side of the Atlantic.
Abraham had three objectives with his presentation:
- Explain his findings on how to improve income in retirement,
- show off his firm’s retirement planning software, and
- plug his book, which was published the week of the conference.
Before we go through Abraham’s slides, let get the other two out of the way.
Here’s the wall chart that comes with TimeLine, to give you some idea of what it’s all about.
Timeline is used to project a client’s situation and plans forward into retirement.
- The wall chart focuses on historical data, but Timeline also supports Monte Carlo analysis.
I haven’t read it yet, but I plan to do so next year.
It’s important to remember that Abraham’s audience was almost entirely made up of IFAs.
- Their priorities and considerations differ significantly from those of the private investor, as we will see when we go through the slides.
These are the levers Abraham has identified to control retirement.
- He doesn’t discuss all of them in this particular presentation.
And these are the ways in which Abraham looks at risk in retirement.
Since IFAs deal with many clients, they need to focus on a range of retirement periods.
- As a private investor, you can just take your life expectancy, and add as many years as you feel comfortable with.
I retired at 50, and my life expectancy is 86, so I tend to use a 40-year retirement.
- This is longer than the standard 30-year period used by the industry until recently.
- But shorter than the 60 years that the extreme FIRE bloggers (such as ERN) are looking at.
Note also that IFAs are interested in success rates (for not running out of money) below 100%.
- A 95% success rate (1 in 20 chance of failure) can be sold to a client.
- I am only interested in 100% success rates (based on historical data).
The legacy measure is interesting, and 50% percentile is probably the position I would choose for myself.
- I have no descendants, but this terminal post provides psychological comfort by acting as a buffer/reserve tank if investment conditions deteriorate.
Portfolio longevity is doubling up for me – I just want to know that I won’t run out of money in 40 years.
Neither am I massively interested in the income measure, but I can see how it might be useful to compare strategies that otherwise seem equal.
Abraham uses historic returns to assess the impact of volatility.
- He uses monthly returns from Jan 1926 to Dec 2017.
- That gives 865 results for 20-year periods, down to 635 results for 40-year periods.
Here’s a chart of UK bull and bear markets:
And here’s a chart to show that bonds have bad years, too:
Here’s a useful comparison between asset classes, showing:
- Average returns pa
- Highest return in a year
- Lowest return
- The proportion of winning and losing years
Abraham’s baseline portfolio is:
- 50% global stocks
- 40% global bonds
- 10% cash.
More interestingly, his baseline withdrawal rate is 4% pa.
- We know from ERN’s work that the highest sustainable (bullet-proof) rate is 3.25%.
The reason for the discrepancy is that IFAs are often dealing with people who want to retire, but might not have quite enough money to do so.
- Upping the withdrawal rate means they have enough money to live on (for a while).
There are a few points here:
- The highest 40-year success rate is 85% (!).
- This was the “no rebalancing, spend cash first” option.
- On that scenario, the worst case was that you ran out of money after only 25 years.
So this baseline portfolio is not safe at 4% pa withdrawals.
To improve things, Abraham changes the asset allocation:
- 40% global stocks
- 30% global bonds
- 10% emerging markets stocks
- 10% gold
- 10% cash
Now the best option produces a 100% success rate for 40 years.
- Note that the worst performance for that portfolio is listed as running out of money after 30 years, so it must be close to 100% rather than exactly 100%.
There are improvements across the board.
Abraham also looks at an aggressive portfolio:
- 50% global equity
- 30% global bonds
- 10% emerging markets
- 10% gold
This switches the 10% of cash in the optimal portfolio into global stocks.
- This is something that I would find very hard to do.
From my perspective, the best possible option has remained the same.
- Some of the lesser options have moved a bit closer to it, though.
5% pa withdrawals
I’ve no idea why anyone would expect to get away with 5% pa withdrawals, so I assume that this section is intended to show how the various strategies degrade under pressure.
Here’s the aggressive (“best”) portfolio:
There’s nothing good enough for me here – the best 40-year success rate is 78%.
Abraham pauses at this point to summarise what we’ve seen so far:
- Equities are king, bonds don’t do much.
- Emerging markets and gold help.
- Cash buffers don’t work, but
- East equities last
- I think a cash buffer has psychological value, especially at the start of retirement.
- And it works as a mini reverse equity glide path if you spend it before your stocks.
Next, Abraham looks at Prime Harvesting (the Michael McClung strategy) vs an equity ceiling.
- These are both rebalancing approaches, where you sell stocks when you have “too many”.
Harvesting sells 20% chunks when the stock allocation gets to 120% of its original value.
- The Ceiling rebalances when above a set figure (eg. 60%).
Unfortunately, Abraham uses the baseline portfolio with 4% pa withdrawals.
- So none of the results are good enough for me.
Comparing the “best” strategies for the baseline portfolio at 4%, all we really learn is that spending your cash first seems to help.
Looking at the strategies for the aggressive portfolio at 5% pa, harvesting seems to help, and you need to spend your bonds first (since you have no cash).
- But none of these options are good enough for me.
After tweaking asset allocation and rebalancing, the next thing to look at is variable spending.
- Abraham picks out four rules:
Here are the results, for the baseline portfolio at 4% pa:
That table is a bid hard to read, so let’s split it in two:
As we’ve already seen, without dynamic spending, these options look grim.
But Guyton’s guardrails appear to fix everything (in terms of safety, at the expense of reducing the total income received during retirement).
Here’s the same table for an aggressive portfolio at 5% pa.
- Note this is a new “extra aggressive” portfolio, with another 10% shifted from bonds into stocks.
Once again, the left hand side of this chart is grim.
And once again guardrails save the day (if you don’t mind lower income overall).
When I saw Abraham’s presentation last year, I thought it was the best thing I’d ever seen on decumulation.
- But now ERN has spoiled me.
I still think there’s some good stuff in there:
- You need a lot of stocks
- EMs and gold are good diversifiers
- Spend your cash first, then your bonds
But a lot of it is not relevant to me as a private investor with a 40-year retirement:
- I know that I can’t be safe with a withdrawal rate above 3.25% pa
- And guardrails only look safe because they dramatically reduce the amount of money that you can take out each year to live on.
At the same time, it’s fascinating to see how the industry looks at the problem.
- And it was a really enjoyable presentation.
Until next time.
Article credit to: https://the7circles.uk/sorc-18-part-4-improving-retirement-income/