How would you build a perpetuity?

jamesd

Triumph13 wrote: »

I've looked at some of these rules and they are definitely interesting, but all the modelled scenarios seem to be against the objective of not running out of money rather than maintaining the value of the pot, with a 'perfect' solution being to pop your clogs with just enough left to pay for the funeral.

They are, but to do it they look to preserve capital, which is one of your objectives.

Triumph13 wrote: »

The key question I've been struggling with is whether to skew the portfolio to income to get the yield naturally, or just have a diversified portfolio and do a mini rebalance one a quarter to get my cash holding to .75% (or 1% if things go well) and draw it out.

If you have a substantial amount in say a global tracker fund it's just some of that to sell. You should probably keep a year of investment income in a savings account to smooth out the results and get the improvement in success rate which has been shown for that approach. Then you won't have much reason to wonder about income variation and could trade every couple or three years instead of annually.

gterr

jamesd wrote: »

Your plan seems reasonable for your objectives. While it's not guaranteed you can reasonably expect to take around 4% of the initial value increasing with inflation and have a low chance of having to substantially decrease the income nearer to end of life.

I'm just wondering how this approach varies (arithmetically or otherwise) from taking a flat 4% of the total (current value) each year, which is what I was planning to do.

jamesd

Choosing to let your income drop with inflation means you're also choosing not to take all of the available investment returns out. That in turn means that the size of your pot is likely to grow throughout retirement and end up substantially larger than the amount you started with. Even in the 2-3% inflation world we've become used to, not taking out the equivalent of 2-3% of the capital value each year is a big deal.

Triumph13

I think what gterr was suggesting was taking a fixed percentage of the current value each year rather than a fixed pound figure based on the starting value.


It is certainly a possible method of taking a broadly sustainable income, but it will be very volatile through the ups and downs of the market. That probably means you need to use larger cash funds outside this main portfolio to smooth the income and you then get into the question of is this the most efficient method when you look at the combined picture.


The approach I was looking at has the advantage that dividend streams are generally less volatile than capital values and so should give a smoother income with consequently less need to tie up cash in smoothing pots.


I think that gterr's approach does have a lot to recommend it, but I think you'd need to do it much more frequently than once a year - say 0.33% once a month rather than 4% once a year - to give a more sensible degree of volatility.