more than rebalancing: keeping it mechanical

Thrugelmir

Tom99 wrote: »

[FONT=Verdana, sans-serif]I don't think that's correct. If a particular share in a tracker goes up in value by say 50%, the tracker does not have to buy any more of that share as their existing holding has gone up in value by 50%. In the same way a tracker does not have to sell a share when its value goes down.[/FONT]

Agreed. To expand further. The constituents of an index do not remain static. There are regular promotions, relegations, floatations and exits. plus rights issues to fund. All of which require a sizable tracker fund to rebalance to maintain it's correct weightings.

Then there's free float to be considered. With many companies shares effectively untradeable. As locked in by family, founder and management holdings.

Trackers aren't as simple as they may seem. MSCI has a complex methodology.

Tom99

james_09 wrote: »

About 5 years ago everyone was saying stocks were massively overvalued and a crash was imminent. If you had followed the original poster's strategy 5 years ago you would have significantly under performed a buy and hold strategy.

Who's to say the next 5 year's won't be the same?

[FONT=Verdana, sans-serif]But the OP's formula does not rely on what everyone is saying about the future only past performance.[/FONT]

short_butt_sweet

Tom99 wrote: »

[FONT=Verdana, sans-serif]I have put some numbers into a FTSE100 spreadsheet to see what results they give but as for additional parameters:[/FONT]

• [FONT=Verdana, sans-serif]What time period? 2010 – 2018 or 1998 -2018[/FONT]

[FONT=Verdana, sans-serif]as long as possible.[/FONT]

• [FONT=Verdana, sans-serif]Dividends, say 3%pa reinvested?[/FONT]

[FONT=Verdana, sans-serif]something like that. though it would be better if you can find total return index figures.
[/FONT]

• [FONT=Verdana, sans-serif]Interest on cash, say 3%?[/FONT]

[FONT=Verdana, sans-serif]well, over the last few years that would be somewhere near. though less accurate further back.[/FONT]
[FONT=Verdana, sans-serif]
[/FONT][FONT=Verdana, sans-serif]i'm not sure what's a good source for realistic long-term interest rates on cash.[/FONT]
[FONT=Verdana, sans-serif]
[/FONT][FONT=Verdana, sans-serif]perhaps bank base rate + 1% would be a decent approximation?
[/FONT]

• [FONT=Verdana, sans-serif]Trigger to buy/sell, say change in index +/- 3% to 10%[/FONT]

[FONT=Verdana, sans-serif]perhaps when current % in equities is 5% out from its target?[/FONT]

[FONT=Verdana, sans-serif]so if we compare (starting with £300,000 capital):
[/FONT]
[FONT=Verdana, sans-serif]a) 50% in equities, 50% in cash, with rebalancing[/FONT]

[FONT=Verdana, sans-serif]b) 75% of the first £150,000*, plus 25% of the remaining capital, [/FONT][FONT=Verdana, sans-serif][FONT=Verdana, sans-serif]in equities; the rest in cash[/FONT][/FONT]

[FONT=Verdana, sans-serif][FONT=Verdana, sans-serif]*initially; this figure to be increased by RPI + 2% over time
[/FONT][/FONT]
[FONT=Verdana, sans-serif][FONT=Verdana, sans-serif]then portfolio (a)'s target for equities is always 50%; you'd rebalance when equities reach 45% or 55% of the total.[/FONT][/FONT]

[FONT=Verdana, sans-serif][FONT=Verdana, sans-serif]and portfolio (b)'s target for equities would start at 50% but move gradually. e.g. if equities rose, the target might have fallen to 49% by the time equities held had risen to 54%, at which point you'd be 5% off target and would rebalance.[/FONT][/FONT]

[FONT=Verdana, sans-serif][FONT=Verdana, sans-serif]i suppose that implies that portfolio (b) would be rebalanced a bit sooner than portfolio (a). would it be a better test instead to rebalance (b) at the same time as rebalancing (a)? i'm not sure.
[/FONT][/FONT]
[FONT=Verdana, sans-serif]

How do you think the formula approach would match your needs and ability to take on risk? It is, after all, just a fixed formula and your needs and risk profile will alter over time and not be based on how the equity market is performing.

[/FONT]
some of my capital is long term. there's no way i'm touching it until i'm much older. it will either support me in my old age, or (if i've been too cautious / lucky in investment returns) be given away or left to other people / charities.

but some of it is shorter term, or more tactical. i'm not talking about capital that will definitely be spent soon - that stays in cash (or similar), and isn't considered for possible investment. this is about capital that might be used to buy a home / upgrade to a more expensive home / buy a second home / let me maintain my lifestyle while doing a less well paid job / etc. some people might keep all of this in cash, or in accounts at fixed rates for 1 to 5 years. but i'm prepared to put a part of this capital in equities, in the hope of at least preserving the real value of the tactical capital. despite the genuine risk that it i might want to use it when the value has fallen.

short_butt_sweet

Thrugelmir wrote: »

Why? Sell your winners and buy more of the duds. Investing should be more than a purely mechanical exercise. Shares do fall for fundamental reasons.

i'm not suggesting this mechanical process be applied to individual shares. only to your overall level of exposure to equities.

short_butt_sweet

james_09 wrote: »

If you think the above strategy is guaranteed to do better than simply buy and hold, why is everyone else not already doing it?
Do you think the original poster has just stumbled across a way to beat the market that everyone else has missed? The answer is this strategy may beat the market, or it may not.

you're certainly correct that it may or may not beat the market. and that's not the whole aim. it's about slightly different kinds of rules which can be used to keep your portfolio in line with your need, ability and willingness to take risk.

one method is to put a suitable % in equities and then just hold what you're got. the % in equities will drift over time, probably upwards.

another is to put a suitable % in equities, and rebalance (after some time or threshold is passed) back to that %.

my suggestion is similar to the latter method, but to over-rebalance. i.e. when equities shoot up, you over-rebalance to slightly less than your original % in equities; and when equities crash, you over-rebalance to slightly more than your original %.

i suspect this is not radically different to simple rebalancing. it's probably just a minor variant on the latter. and one thing to be said against it is that it's more complicated.

it's also not new, in that i've seen some mention of the idea of over-rebalancing. though i haven't seen a formula for it, i.e. a way of making it truly mechanical.

masonic

So past performance is no guide to the future, but a little backtesting can be helpful when thinking through an investment strategy.

Let's assume you want to execute on this over-rebalancing strategy and for sake of argument you start at the bottom of the global financial crisis at 100% equities, and you make your over-rebalancing decision each year on 1st January.

Over the 10 years 2008-2018 (where the FTSE World index has risen a little over 250%), what percentage equities would you have ended up at and what effect would over-rebalancing have had on your returns? Hopefully you aren't at less than 0% equities, because that would be impossible. Consider this step a calibration.

Next you could apply exactly the same rules to the prior 10 years (1998-2008), when there was only a 60% rise, but plenty of ups and downs. Those seem like prime conditions for your strategy to deliver great results. So does it?

The market data is all available via Trustnet advanced charting.

AnotherJoe

A simple question "define crash"
A second simple question, referencing James post at 11, which I'll rephrase as "why do you think you're so clever you have spotted an algorithm that literally billions of dollars worth of computing power and ultra sophisticated algorithms haven't already worked out" ?
Or to paraphrase HL Mencken "for every complex problem there is a simple solution..... which is wrong"

Thrugelmir

short_butt_sweet wrote: »

i'm not suggesting this mechanical process be applied to individual shares. only to your overall level of exposure to equities.

Equities can take different forms. Which in themselves provide diversification in these correlated times.

Thrugelmir

AnotherJoe wrote: »

A second simple question, referencing James post at 11, which I'll rephrase as "why do you think you're so clever you have spotted an algorithm that literally billions of dollars worth of computing power and ultra sophisticated algorithms haven't already worked out" ?

Even the rocket scientists employed by the hedge funds years back. Omitted to build a "zero" eventuality into their must win market roulette wheels.

Tom99

[FONT=Verdana, sans-serif]I have put your alternative rebalancing options into a spreadsheet using the FTSE100 Index for the past 20yrs since 1998:[/FONT]

[FONT=Verdana, sans-serif]Option 1 : Invest £150,000 in shares £150,000 in cash and leave alone[/FONT]
[FONT=Verdana, sans-serif]Option 2 : Rebalance to 50% shares and 50% cash[/FONT]
[FONT=Verdana, sans-serif]Option 3 : Rebalance to 75%*£150k(linked to RPI+2%) plus 25% of remainder[/FONT]

[FONT=Verdana, sans-serif]I made the following additional assumptions:[/FONT]

[FONT=Verdana, sans-serif]Dividends reinvested: 3%pa (I can't find any actual data)[/FONT]
[FONT=Verdana, sans-serif]Cash Interest: Base Rate +1%[/FONT]
[FONT=Verdana, sans-serif]Trigger to rebalance: When share %age differs by more than 5% from target[/FONT]

[FONT=Verdana, sans-serif]The result are:[/FONT]

[FONT=Verdana, sans-serif]........................Option 1... Option 2... Option3[/FONT]
[FONT=Verdana, sans-serif]Start: .............. £300,000.. £300,000.. £300,000[/FONT]
[FONT=Verdana, sans-serif]Share Profit:...... £228,833.. £288,003.. £306,097[/FONT]
[FONT=Verdana, sans-serif]Cash Interest:... £186,398.. £176,940.. £186,561[/FONT]
[FONT=Verdana, sans-serif]End: ................ £715,231.. £764,943.. £792,658[/FONT]
[FONT=Verdana, sans-serif].........................+138%.......+155%.... +164%[/FONT]

[FONT=Verdana, sans-serif]The Buy/Sell deals:[/FONT]

[FONT=Verdana, sans-serif]............................Option 2... Option3[/FONT]
[FONT=Verdana, sans-serif]Buy deals: ................. 6 ............ 17[/FONT]
[FONT=Verdana, sans-serif]Sell deals:.................. 7 .............20[/FONT]
[FONT=Verdana, sans-serif]Min Buy: ............... £18,023.....£15,865[/FONT]
[FONT=Verdana, sans-serif]Max Buy: .............. £28,975.....£41,254[/FONT]
[FONT=Verdana, sans-serif]Buy Avge:.............. £23,010.... £26,043 [/FONT]
[FONT=Verdana, sans-serif]Min Sell:............... -£19,535... -£16,571[/FONT]
[FONT=Verdana, sans-serif]Max Sell:.............. -£37,642... -£40,921[/FONT]
[FONT=Verdana, sans-serif]Sell Avge:............. -£27,173... -£26,553[/FONT]

[FONT=Verdana, sans-serif]A lot of the extra Option 3 deals are in 2008 (10 deals in 2008)[/FONT]

[FONT=Verdana, sans-serif]I would add the following cautions:[/FONT]

[FONT=Verdana, sans-serif]1 – My maths, are they correct? Anyone care to check?[/FONT]
[FONT=Verdana, sans-serif]2 – The future will not replicate the past.[/FONT]
[FONT=Verdana, sans-serif]3 – A different time frame or index would produce a different result/..[/FONT]
[FONT=Verdana, sans-serif]4 – No Fund, Platform or dealing costs allowed for.[/FONT]
[FONT=Verdana, sans-serif]5 – The 3% Div and cash interest at Base +1% may be flaky estimates.[/FONT]