However since the last time i was on TPAW, i would have been better off , just being in 2 funds. WTF.
Shows what i know.

As far as I can make out, reading Ben Mathew's posts, I think it is a complex interaction between:-

a) the expected equity risk premium (expected to be relatively low in the future).
b) what fixed income streams you have in the future, e.g. SP, DB, etc.
c) whether you have specified any "essential expenses", which TPAW forces to come from bonds.
d) maybe other things I've missed ! (^_-)

It's worth reading the Bogleheads TPAW thread slowly, from start to finish, to at least get a feel for some of these concepts. Ben tries hard to explain it simply for us lesser mortals, and his insights are quite profound, and sometimes counter-intuitive... If you share your plan with him, he might even go through it line-by-line.

SWR has misled millions of people over the years, into enduring sub-optimal (and risky) retirements.

The Bengen study didn't even recommend it ! It just seems to have been adopted by lazy, innumerate "financial advisors" en masse... [admittedly, Bengen predated the internet and the advent of powerful personal computers/spreadsheets]

Merton identified the amortization and "lifecycle" approach as long ago as 1969 as being optimal, but it is only comparatively recently that it has been taken up and reached a wider audience, by the likes of VPW and TPAW using spreadsheets and online tools.

Something I can't understand about the TPAW tool is that, despite leaving the controls at the default position, the 50th percentile withdrawal level gradually increases (in real terms) over the 30 year retirement period.

I would expect the opposite to apply; spending in retirement should decrease through the so-called go-go; slow-go and finally no-go years, or at the very least decline gradually.

The tool does allow you to tilt withdrawals in either direction, but what is the rationale for the default?

Thanks for the post, an interesting approach.

How sensitive is it to the R parameter? I'd be interested to know what it produces for R=0 (presumably a future with zero real growth).

And at the risk of pushing my luck, would a stagflation scenario be modelled by inputting a negative value for R?

One thing I think tends to be missing in these models is health risk/expectation - I'm probably going to be front-loading my retirement income to spend more of it earlier. The logic being you can enjoy it more when you're 60 than when you're over 80 and you never know when your mobility, eyesight, etc, might start tailing off. So rather than the 4% rule I might do the 6, 5, 4, 3, 2 rule as I'm gradually moving towards my date with the grim reaper.

Yes, you can set almost any glidepath you want.

Here's what I intend mine to look like (somewhat, I haven't finished tinkering!) I intend to officially "retire" next year, aged 60. [I've been effectively retired since age 35, but hey-ho...]

Basically, I want to extract the most now, while still getting at least £40k pa (at the 5th percentile) in the unlikely event I should live towards 90...



There are numerous stocks and bonds return forecasts you can try in TPAW. Under the "Fixed" option you can reduce both to -1% annual real returns, if you wish, which may be what you're thinking of.

It is important to enter all your fixed incomes expected during retirement, e.g. SP, DB, etc. to give you a true picture. [e.g. full UK SP is currently £958 a month, expected to outpace inflation]

Another reason why SWR is crap in comparison.