RetireJapan

This old chestnut. I understand the role of bonds in diversification and that people are rethinking the old ‘age-10’ allocation to make it a smaller proportion of the entire portfolio. As I near retirement I’m in a mind to start increasing my allocation. Our iDeCos are already over 25% in an index fund and now I’m thinking about putting some in my NISA. Except you can’t buy bond indexes in NISA. So I’d have to buy in our tokutei accounts. Seeing as how they’ve hardly over-performed in recent years, I’m wondering since I’ll have to pay the 20% CGT, is it really worth allocating more? I wonder what other folks views are (I realize there’s no right or wrong answer). Retirement will be within 10 years.

banders wrote: ↑Fri Jan 31, 2025 10:18 am... Seeing as how they’ve hardly over-performed in recent years, ...

Ha! (Maybe bonds will stage a miraculous comeback?)

Just as a data point, I'm 73 and am still all equities/no bonds, tho with an oversized bucket 1 of cash on the side. (There are a few other personal quirks to our situation that make this approach not so worrisome.)

I retired a couple of years ago, and as I will start drawing my pension later this year, I am in the process of rebalancing my portfolio(s) to a more bond heavy allocation.

At the moment I'm maybe 45% bonds + 55% stocks, but I'm aiming to get this to around 60% bonds + 40% stocks taking my cue from the allocation of Vanguard UK's Target Retirement Fund which goes from around 40% bonds + 60% stocks when the investor is 10 years from retirement, then to around 45% bonds + 55% stocks when the investor has just retired, and then to around 60% bonds + 40% stocks 5 years into retirement.

Portfolio Statement (as at 31 March 2024) for the Vanguard Target Retirement 2020 Fund


One interesting point about the allocation fine tuned by Vanguard's fund managers is the over-weighting to UK domestic stocks and bonds. Although UK stocks currently make up only around 3.4% of the FTSE Global All Cap Index, around 1/4 of the stock allocation in the Target Retirement Fund is comprised of UK stocks, and when it comes to bonds, around 1/3 of the bond allocation of the Target Retirement Fund is made up of UK bonds (with a substantial chunk of the bonds allocation changed to inflation linked gilts when the investor is 5 years into retirement).

The home country bias obviously has the benefit of reducing currency exchange rate risk in the retiree's portfolio.

Since the BOJ is still in the process of increasing their short term interest rates, I am currently sitting on the fence as far as buying seriously into JGBs, or increasing my international bond holdings since the yen is still relatively weak. Since the Vanguard fund uses only GBP hedged international bond funds in their allocation, I too am planning to shift my international bond holdings to JPY hedged ones maybe next year when the interest rate gap between the US and Japan will hopefully have narrowed and currency hedging costs are more reasonable. In the meantime, I am quite heavily into unhedged US bonds and international bond ETFs which are USD hedged (not JPY hedged which is what I really need).

banders wrote: ↑Fri Jan 31, 2025 10:18 am As I near retirement I’m in a mind to start increasing my allocation. Our iDeCos are already over 25% in an index fund and now I’m thinking about putting some in my NISA. Except you can’t buy bond indexes in NISA.

Maybe not in the tsumitate but you can in the growth portion.

https://www.rakuten-sec.co.jp/web/fund/ ... 90C000END3

An interesting article from Andrew Hallam on a scenario where stock and bond porffolio will outlast stock only
https://www.aesinternational.com/blog/h ... s_amp=true

Thanks for the replies, guys. That Hallam article was an eye-opener! I think I'm going to buy the bond index for the next couple of years to bring it up to 60/40.

Hallam is talking about sequence of returns, and combining that with a couple other presumptions. And I do realize there are clear and possible dangers to a person's money lasting long enough in retirement.

One presumption is that the bad sequence of returns will happen the year you retire (worst case scenario), and perhaps (necessarily?) continue for several years. I have read Hallam's article and others that explain/describe this possibility, but nobody (to my knowledge) has taken it a step further to look at the odds of that happening. If I were to do so, I'd look back maybe 30yrs to 1995 and hypothesize someone with a million dollar nest egg retiring in each successive year, up to about 2020 (would still be able to see if a 2020 retiree would have hit a bad sequence right off, or in the first five years of retirement).

There were two bad sequences in this 30yr period. Hallam uses the 2000-2005 period, a textbook example, tho you could look farther back to the early/mid 70s, or 11/2007 to 11/2012 for something similar. There are some other 'crashes' or blips (or whatever they might be called) that I don't think would create a bad sequence, or certainly not as bad--there was the crash of '87 but the market recovered comparatively fast; in the early 90s the market kind of had the blahs for a couple years; the crash at the onset of covid (shortest blip here?); and maybe calendar 2022 and most of 2023 could be an edge case. And as with anything, when studying this, you'd have to make some judgements: eg, is a bad sequence of returns necessarily a three year span, or is two enough to cause a retiree's saving to run out? Or, does a bad span of years necessarily also include a certain percentage decline, and if so, how much decline does there need to be to tip the scales into disaster? If I could I'd like to ask Hallam to run the same numbers on 2007-2012 to compare results--how would each of those charts look for this more recent time period?

And so on.

Another presumption (two?) in this type of article is that the nest egg figure ($1 million) is otherwise just the right/exact number for a retiree using the 4% rule. What about if someone has over-saved? How do the numbers work if a person has $1.5m, or 2m? (and still takes the $40k/yr associated with a million dollar figure) I'm pretty sure the picture will be far less dire. Or, change the other variable. Stick with the $1m and use a 3% (or 2%) withdrawal rate--and again, I think this hypothetical retiree (all in stocks, no bonds) would be okay.

One aspect of the bucket approach might also be relevant. Bucket 3 (ten years out, or more) is usually framed as being 100% equities. I think the reasoning on this is that over any given 10yr period, equities virtually never lose money. Worst case might be that after ten years the value would be the same as at the start (a bad sequence, followed by enough of a recovery to only reachieve par). There are definitely articles/research on this aspect of things.

On the positive side, sometimes a sequence of returns can be wonderful. I'm not going to search out any concrete numbers, but in yen terms, from 2019 to the present, a dollar-based 100% equity portfolio (without any additions or withdrawals) would now be up about 2.5 times. So if that portfolio was equal to ¥10m in 2019, it'd be worth ¥25m now. And that span includes the covid crash, and the mentioned 2022-2023 period. As in that old movie title, the gods must be crazy.

*

In other news (personal), on Friday one of our kids got a full pass on their thesis defense. Yay!

banders wrote: ↑Sat Feb 01, 2025 2:05 pm Thanks for the replies, guys. That Hallam article was an eye-opener! I think I'm going to buy the bond index for the next couple of years to bring it up to 60/40.

I think that it depends on how much risk you’re willing to take in retirement. Me, I think that catastrophes can hit us out of the blue, especially with the biggest green house gas emitter nation in the world gearing up to drill baby drill…

This Bogleheads discussion about the paper “The Safe Withdrawal Rate: Evidence from a Broad Sample of Developed Markets” (Arnakulova et al, 2022) might interest you too.

ChapInTokyo wrote: ↑Sun Feb 02, 2025 2:14 am Me, I think that catastrophes can hit us out of the blue, especially with the biggest green house gas emitter nation in the world gearing up to drill baby drill…

China is the biggest emitter is it not?

sutebayashi wrote: ↑Sun Feb 02, 2025 12:06 pm

ChapInTokyo wrote: ↑Sun Feb 02, 2025 2:14 am Me, I think that catastrophes can hit us out of the blue, especially with the biggest green house gas emitter nation in the world gearing up to drill baby drill…

China is the biggest emitter is it not?

Nice video here

https://youtu.be/S1hJnUk_Mbc