Warren Buffett : How to spend $145billion

Successful companies make money and hand some of it back to their shareholders via dividends. Some of it they may keep and this piles up over the years. It can be quite a sum. For Ford it's close to $25billion. For Coca Cola it's past $60 billion.  For Berkshire Hathaway its $145 billion. (The last accounts say $143.7 billion  but that was months ago, its piling up at a £1billion plus every month).

No wonder warren Buffett is so good at allocating capital... he's had much more practice than anyone else on the planet.

Blogger owns a bit of Berkshire..... this is for investment purposes only

Buffett bites back

Buffett has a remarkable record with his chief executives. He has hired good ones, kept them in place, and praised them lavishly even if the company is faltering (in his view good managers in bad businesses are still good managers). Until a few years ago he'd never been rude about them, let alone fired one. An extraordinary record over several decades. And then came NetJets, and the news the chief executive had been let go and a new one installed. In his latest 2011 report Buffett is able, once again, to lavish praise on his NetJets chief executive (now Jordan Hansall).

"A few years ago NetJets was my number one worry: Its costs were far out of line with revenues, and cash was hemorrhaging. Without Berkshire’s support, NetJets would have gone broke. These problems are behind us, and Jordan is now delivering steady profits..."

In Buffett terms this is a rare, and stinging, rebuke to earlier management failures.

Disclosure:blogger owns (a bit of) Berkshire Hathaway

Brits power Buffett profits

Within the Berkshire Hathaway's end of year accounts (out this weekend) there is a breakdown of their regulated utilities businesses (called MidAmerican). Despite the name, it include a large chunk of British assets (Northern Electric and Yorkshire Electricity). Operating earnings (before interest and taxes) for the division are up $120M, but all of that is accounted for by UK utilities earnings (up $136M).

So it seems the regulator is keeping a pretty tight grip on the profitability of these regulated industries, with one exception.

Notes

See: http://www.berkshirehathaway.com/2011ar/2011ar.pdf

(full disclosure: this blogger owns Berkshire Hathaway shares)

2012 : The Year of the Churn

In China next year will be the Year of the Dragon - in the UK it looks set to be The Year of the Churn.

The reason? From 2013 IFA's will be banned from taking commission from clients, except there is a loophole as watchdog Consumerfocus notes.

It is particularly galling that existing customers will continue to pay trail commission on their newly written contracts – introduced just before the ban – potentially for several decades.

So there's an immediate incentive to get clients into contracts - or if they are already in contracts to switch them to higher paying one. And this process may may have already started - there's evidence of a pick up in activity in these areas.

Commission based advisors usually get paid a lump sum from companies they steer their clients into, plus an ongoing 'trail commission' of around 0.5% a year. Quite what this is for is unclear. Some see it as deferred fee, others as for 'continuing service'. And it can go on for years and years.

The RDR (retail distribution review) will end trail commission - but in the short term it may have increased it.

There's something weird about Warren

Warren Buffet's annual letter to share holders is out today and well worth the time spent reading it. (http://www.berkshirehathaway.com/letters/2010ltr.pdf).

Well the number one weird thing is the share price. $100000 plus. And part of the reason for that is the way Warren Buffet measures Berkshire Hathaway's (ie his) performance.

You won't find net income given much shrift as a useful figure - in fact it's booted into the long grass. But Warren's favourite metric - growth in annual per share book value - now has a competitor. The rolling five year return.

From 1965-2010 there are 42 five-year periods. Berkshire Hathaway beat the S&P in every period. This is a stellar performance. Not only that: the S&P had six negative periods in that time (nominally at least, the after-inflation picture would be less rosy). Berkshire had none. It was safer and more profitable than the index.

How did they do it?

Here's one quote from the letter:

'At Berkshire, managers can focus on running their businesses: They are not subjected to meetings at headquarters nor financing worries nor Wall Street harassment. They simply get a letter from me every two years and call me when they wish.'

They don't teach that at Harvard.

Pensions: the 10% cashback

So in the end it comes to this for the rich HRT (40% tax) payer: stick £1.5million into your pension fund, and the government gives you 10% - that's £150,000.

It used to be done in the name of pension planning, but under the new rules (from April2011), there needn't be any pension involved. As long as you meet the £20 000 pa minimum income requirement you can take all your cash out of your money-purchase pension scheme at age 55.

25% of that is tax free, and the 40% tax on that £375000 would have been £150000...... The figures work the same if you have £15,000 in your SIPP (self-invested pension plan), but that pensioner may struggle to meet the £20,000 pa income requirement, or pay 40% tax. And while £1500 is welcome, it is scarcely life changing.

it's true that even under the old rules, the 25% tax free cash effectively knocked the 40% tax payer into the 30% bracket. But now the process is more transparent, the income stream (aka pension) can be avoided altogether and the whole lot taken as a lump sum. Whether the rich need to be subsided to the tune of £150,000 under pension laws when there is no pension at the end of the process is a question well worth asking.

Jam today or jam tomorrow?

OK, any jam right now is wishful thinking. But what should you wish for? There is an answer to this, and it depends on who you are. Savers want one thing, and pensioners another. Because savers are accumulators, and pensioners de-accumulators. Both want maximum return at the moment they hold the maximum amount of money - which is at retirement. So if you faced 7lean years, and 7good years, savers want the lean years first. So its jam tomorrow for them.

For pensioners - as they run down their stash - they want the good years first. So for them it is actually jam yesterday. They want the best years to have happened at retirement.

But what about someone who invests a fixed lump sum and leaves it untouched? For them the order of returns doesn't matter.

Which is odd, because it means no one really needs jam today.

A pension on £10 a day

The simple pension question 'How much will I get?' is rarely answered succinctly.

If you save £3650 a year for 40 years, you will end up with £250 000 (assuming 2.5% interest rate).

£10 a day gets you £10 000 a year at 65

How reasonable are these assumptions? The cost for the same product in 2050 is unknown - if interest rates rise you could retire earlier. If lifespan increases you might have to wait until you are 69 to get the same pension. But right now a 65year old man can get an index-linked £10000 a year for £250 000.

Interest rates? The vanilla benchmarks are 1% for safe and 4% for risky assets. - gets you to 2.5% for 50/50 mix. And 40 years will take you from 25-65. So not completely unreasonable. By the way this a 'real return' benchmark. Your '£250 000' in 2050 will be worth £250 000 in todays money - even though the cheque might actually say '£1million'.

Given UK median income of around £25000 then the savings rate is about two months income per year.

So stashing £300 a month into an ISA can be a simple sensible option. There are far more complex options available - but they are not necessarily any better.

Factor30 and the 10k pension

The warming glow of low interest rates is intended to make the economy bloom. It makes pensions shrivel.

If you have a defined benefit pension you can skip the rest of this. You don't have to pay cash for your pension. But those on defined contributions (aka a pot of pensions savings) do. A 60year old women now needs to reach for Factor30 before heading to that retirement beach. Factor30 means she will have to multiply her desired pension (aka annuity) by thirty to get the cost.

£10 000 RPI linked pension at 60? That will cost you £300 000 madam.

Men have it easier (because they don't live as long), but a return of 3.35% for men aged 60 v 3.2% for women is nothing to write home about.

What to do? Should she hand over her £300 000 now? The choice is even more varied because the new government says you won't have to buy an annuity by 75 (actually you didn't have to before, but the alternative was so unattractive few people chose it).

First off our mythical 60year old pensioner - lets call her Alice - can increase her pension by 75%. All she has to do is forget index linking and go for a flat rate. The return jumps from 3.2% to 5.55%. For handing over her £300 000 she now she will have a fixed £1400 a month for life.

She could wait another five years - the current RPI-matching rate for 65year olds is 3.75%. Or she could continue to invest her pension pot and drawdown an income from it.

In an earlier blog we came up with a rule of thumb for drawing down from your savings: 4% a year (of that years total), rising to 5% in you sixties and 6% in your seventies. It meant your annual income would jump around as share prices went up and down, but there are ways of smoothing it.

So lets look at the various returns Alice could get on her money from annuities and/or drawdown.

Return rate 3.2% 4.7% 5.5% 5.0% from drawdown

Growth rate RPI 3% 0%/fixed ?/income could drop

income* £800 £930 £1400 £1250

*(monthly from £300,000)

I have thrown in an additional option of 3% growth. This may be better value than pure RPI, because the insurance companies at least have the certainty of knowing what their future payments are. Companies don't like uncertainty, and it's the consumer who ends up having to pay a hefty premium if they want the absolute certainty of RPI indexing.

What should Alice do? There is no single right answer here. Let's assume she needs £800/month immediately, and she has a pension pot of £300 000.

She could get £800/month by paying out all her £300 000 pension savings for RPI protection, or by paying out £175 000 to get a flat rate.

Alice makes her mind up: half annuity, half drawdown. She pays out around £80,000 for a flat rate annuity that gives her £400/month, and she draws down £400/month from her remaining savings (£220 000). Thats a drawdown rate of 2.2%.

Under this plan her monthly income will decline as inflation kicks in. But she still has her £220 000 fund to draw money from . How she should invest that £220 000 is another question.

note: go to calculators section at

http://www.moneymadeclear.org.uk/hubs/home_pensions.html

to check out annuity prices. Its not a complete list but it is a useful indicator. The rates quoted above are for a 'vanilla' pensioner: non smoker with no kids/spouses benefits, and no benefits/payback if you die the next day.

The Graham formula

Benjamin Graham wrote a 700 page book on security analysis - so two paras from his November 1963 talk may not do him justice.

'Buy definitely 'bargain issues'. Typically these would be shares that sold for less (than) their value in working capital alone, with nothing paid for fixed assets or goodwill. These were quite numerous up to as late as 1957...'

Looking at the market as a whole, Graham valued it based on:

'average earnings of the 30 stocks in the Dow Jones industrial average for ten years past, capitalised at twice the interest rate on high grade bonds. For example at the present time the average earnings for the last ten years are about $33 on the Dow Jones unit and the present value on high grade bonds is 4.3%. If you capitalise at 8.6% - which is a multiplier of about 12 - you would get a Central Value on the old basis of about 380, as compared with the present price of about 750'.

Never mind the forecast, feel the philosophy.

Benjamin Graham also says

I was about to chuck out the Benjamin Graham lecture notes - I mean who keeps 47year old typescripts , even if they did originally cost 50c?

But in a handful of pages Graham not only discussed the weight of asset classes in of a portfolio [risky v nonrisky, see last blog], he also looked at what you should buy and when to rebalance.

'Use a 50-50 division between stock and bonds. When the market level of the stocks rises to a point where they constitute 55% of the total or maybe 60%, you would then sell out enough to bring your proportion back to 50% putting the proceeds, back into bonds or saving banks.'

Graham goes on to discuss various similar 'formula plans' -

'The main need here is for the investor to select some rule which seems to be suitable for his point of view, one which will keep him out of mischief and one, I insist, which will always maintain some interest in common stocks regardless of how high the market level goes'

Graham's point being that if you missed out on a bubble (my word) your disappointment (his word) would have been so great -

'as probably to ruin you from the standpoint of intelligent investing for the rest of you life.'

A final word on stock selection:

'there is no indication that the investor can do better than the market averages by making his own selection or by taking expert advice.'

When Graham wrote this in 1963 the first index funds were years away.

Benjamin Graham says

Benjamin Graham was a professional investor, wrote Security Analysis and taught Warren Buffett. His book is heavy going for the uninterested, so how to sum up Graham's advice for the nonreader? Luckily Graham did it himself. I am looking at a typescript of a talk he gave in 1963 (literally a typescript, no powerpoint in those days)

In my nearly fifty years of experience* on Wall St I've found that I know less and less about what the stock market is going to do but I know more and more about what investors ought to do..... the investor is required by the very insecurity ruling in the world of today to maintain at all times some division of his funds between bonds and stocks (cash and various types of interest-bearing deposits may be viewed as bond equivalents). My suggestions is that the minimum position of this portfolio is 25% and the maximum should be 75%.

For comparison the suggested range for the vanilla investor is from 20% equities up to 67% equities (the 'bouncy' bit of the portfolio); giving a range on the 'boring' side (gilts/cash etc) from 80% to 33%. The Graham prescription is riskier (but potentially more rewarding) but even on the 'bond' side he urges caution:

nobody can assume he can get exactly the same degree of safety and dependability in [an] investment yielding 5% as in one yielding 4%.

Wise words as the junk bond era was still to come.

*His career spanned some turbulent times:

I would like to point out that the last time I made any stock market predictions was in the year 1914 when my firm judged me qualified to write their daily market letter

The lecture notes originally cost 50c but you can get them, free, here: http://www.jasonzweig.com/documents/BG_speech_SF1963.pdf

In the mix: annuities, bonds and SIPPs

Let's change the numbers for once. Anna has a SIPP of £133,333. Now she is 60 she would like to draw a pension from it. She is a cautious investor (the C portfolio is 75% boring, 25% bouncy; in her case index linked gilts and global equities).

She can't decide between keeping it invested and drawing an income, or buying an annuity. Access or stability? That is the choice she is faced with. So she goes for a bit of both.

She figures that her investing in index-linked gilts is not that much different from buying an annuity. So if 75% of her SIPP was going to be in bonds, why not just buy an annuity anyway? Sure she says goodbye to the money, but it will give her that certainty she is looking for.

So Anna takes her 25% tax free cash - and pays £100 000 for an annuity. But she is not done yet. She is still earning and saving (and has £33,333 in cash in the bank). So she starts another SIPP and starts paying into that. And, thanks to her regular annuity income, she thinks: 'maybe I am not such a cautious investor any more'.

The moral of this story is that once you reach pensionable age (55) you have the option of swopping the 'bonds' side of your portfolio for an annuity.

Remember a 'pure' annuity actually returns all your money to you over your expected lifetime, meanwhile investing it in gilts. Hence the low-return on your 'investment'. ( A real world annuity does much the same, except it keeps back 10-15% for profits/expenses and holds corporate bonds too).

So annuities have two big drawbacks:

• you lose money if you die early
• your savings are invested at low yields (gilts), possibly for decades

but if your savings were invested in gilts anyway, one of the drawbacks is eliminated.

Anna takes one more decision - she decides to go for the maximum, flat rate annuity. (£475 a month, versus £260 for RPI linked).This 'front loads' her payments from the insurance company and it will help her save still more into her SIPP. And with that income secured she feels she can be a bit more adventurous in her new portfolio.

Vanilla smoothie

Despite the vanilla approach - keeping things as simple as possible - choosing a withdrawal rate from your pension pot is tricky. You can see why most people don't bother trying, and just opt for a standard annuity. It's fixed, it's guaranteed, it will last as long as you live.

But here's ways to smooth your usp (aka unsecured pension, or income drawdown).

1 Invest in government gilts in your SIPP (self-invested pension scheme) or even index-linked gilts. This is a good choice for someone who wants a secure and stable return, but is not ready to lose control of their savings by handing them over to an insurance company for an annuity. This is an option that doesn't get much publicity. The returns are likely to be unspectacular - as are the fees to advisors.

2 Smooth your return by taking the average of last year's payout, and 4% of this years fund total. So if your fund is £115,000 - and last years pay out was £4000- you only pay yourself £4300 and not the £4600 that the basic 4% rule would have given you.

And because last years payout was calculated the same way this simple rule weights your return by your long-run return. (The weighting for the current year is 1/2, and for the preceding years 1/4, 1/8, 1/16 etc).

3 Increase payout with age: 4% in your fifties, 5% in your sixties, 6% in your seventies.

By the time you get to your mid sixties, this is quite conservative (especially for men - their life expectancy is several years shorter than for women). But it might feel scary - your pension pot might be going down. This is not just living on income - it's eating into capital too

So is a SIPP preferable to an annuity? It comes down to personal choice - but one big factor is that you can always change your mind and buy an annuity - but you can't change it back again.

Three kinds of 4%

Nowadays you don't have to buy an annuity with your pension pot - you can invest it and draw an income instead [often known as income drawdown]. This pot can still be used to buy an an annuity later [up to age 75] if you want.

So how much can you safely withdraw each year? Well the answer depends on your life expectancy and your future investment returns. Both are very uncertain, so much attention has been given to finding the 'safe withdrawal rate' even allowing for poor stock market returns, and increased longevity.

The standard answer is around 4%.

But there are three kinds of 4% withdrawal on - say - a $100,000 pot. (We are talking dollars here because its in the USA that this has been studied most).

1 Take $4000 a year index linked. So it goes up with inflation.

2 Take $4000 every year - flat rate.

3 Take out 4% of your pot whatever its size.

Option 1 is the toughest to maintain, option 2 is usually what is meant by the 4% rule, option 3 is the 'safest'. Where safety means you will never run out of money (although you might run short of money).

The vanilla pensioner goes for option 3, despite its drawbacks which are:

• your income will bounce up and down with the value of your fund; and
• it doesn't allow for the fact that as you get older, you should be able to withdraw a higher percentage from your fund

We will look at ways to fix this, without straying too far from the underlying philosophy: the size of the pension you can take depends on the amount of money you have - not on the amount of money you used to have.

And that is the serious drawback of options 1 and 2 - blithely taking $4000 a year even though your fund may have dropped to $60 000.

Will half a million cover it?

This article puts the cost of retirement for a pensioner couple around £500 00 (£600 000 in London).

the average household needs £564,227 to cover the cost of the first 20 years after quitting the workforce. The calculation is based on annual household expenditure for those aged 65 to 74 at £23,107.

Quite what happens in year 21 of retirement they don't say. You'd think they'd know, especially as they can calculate expenses over 20 years to the nearest pound...

http://uk.finance.yahoo.com/news/couples-must-budget-600-000-for-retirement-tele-78e6abbbac10.html

Sweet seventeen

Think of a number and multiply it by 25. Thats a rough guide to the cost of an inflation linked pension. For a woman in her mid sixties, or a man a couple of years younger. So £10 000 pension will cost you £250 000.

Or if you want to retire on two-thirds of your salary, then multiply your salary by 17. That's how much you will need to save.

So what does it take to save an amount equal to 17x your salary? There's a lot of variables here - but it shouldn't put you off working out a ball park figure.

Let start by assuming you save two months of your salary, every year for forty years. And your salary stays fixed in real terms. Take real annual growth as between 1-4%: 1% definitely, 2.5% probably, or 4% possibly (with a lot of your savings in riskier shares).

What do you end up with?

Amount of salary saved after 40 years with growth of

1% 2.5% 4%

8x 11x 16x

So despite continual real growth of 4% for forty years and saving two months salary every year, you still come up short. Just.

Let's up the amount of savings to three months a year.

1% 2.5% 4%

12x 17x 24x

And suddenly we are there, even with only medium growth. With the high returns (4%) you can practically retire on a pension equal to your salary. (It will certainly feel like it, because you won't be saving 25% of your income any more)

But who saves three months money every year?

Shouldn't the government do something? Well they have - they have set the contributions level for the new default pension scheme (now called Nest) at around one months salary [employees pay half of it].

How much would that give you?

1% 2.5% 4%

4x 5.5x 8x

Saving one month's salary a year isn't enough. But the government obviously sees this as a glass half full. Tell people they need to save at least double that amount and they might become too dispirited to even try.

Bolton and vanilla

Hot fund in a hot sector. A friend asks 'I was thinking of putting money into Bolton's new China fund....'

I guess this is how most people begin. Not with boring, but sensible, asset allocation. But a punt on something a bit more exciting. He knows very little about investing. Is this really a good place to start? Surely it would be better to start by having half your savings in global equities and half in index linked gilts? Simpler and less risk.

But the reality is: savers don't start out as vanilla investors - and then move onto more complex savings instruments; they actually move into vanilla investing after having been stung, burnt, or just plain caught by surprise by the vagaries of real world investing.

Back to Bolton. The attraction is his fantastic track record at his Special Situations fund he ran from 1979. Hence the hurry to get into his new fund launching now. 'What do you think?' Hmm. Where to start.

First off: the three big savings choices are house [pay off mortgage]; ISA and pension [company plan or SIPP].

Then you can figure out how much you want to risk in global equities. Then you can have some emerging markets. A subset of that are the BRICs [Brazil, Russia, India China] - and finally we get to China on its own. 'So you don't think I should put all my money in it?' No.

'But the offer closes in a few weeks'. Well, you can still but the shares after the offer closes. 'Really. I didn't know that'. Yup, and then you can buy as little as you like - instead of being forced to buy a minimum of £2500. Or you could wait ten years and buy some then. 'Ten years! The people who bought Bolton's first fund from the beginning did fantastically well'. Yes, it did fine for the first ten years, but it was only in the early nineties, around 1992 that it really took off. And the first five years of its life gave little sign of the fireworks to come. Here's the picture.

'Well I can't risk that - Bolton may only hang around for a couple of years.' Some people would see that as a reason for caution. Besides, you may be able to pick the shares up at a discount.

'Discount?'

Ok, we'll deal with the discount later. Meanwhile I couldn't interest you in a global equites tracker?

'Nah. Too boring.'

Financial journalism

In today's Sunday Times

.... the American-educated prince [Alwaleed bin Talal] - described by Time magazine as 'the Arabian Warren Buffett' lives in a£180M palace with 317 rooms, 250 television sets and gold plated taps.

So, just like Warren then.

Buy losers, sell winners

Vanilla finance has a few very simple rules. They have been shown to work pretty well in the past. But they do have some big implications. 

Remember, the vanilla investor has to pick a mix containing both exciting [eg shares] and dependable investments [gilts]. Boring or bouncy? Whatever lets you sleep easy at night. 

Say you pick a 50:50 mix [the balanced portfolio]. When it gets out of line you will need to rebalance - maybe every couple of years. This will mean selling the asset that has outperformed, and buying more of the underperforming one. You sell winners and buy losers. 

Now buying into a losing share can be a recipe for disaster - because it could go to zero. (Actually this is true of winning shares too, its why we skip holding individual shares  in vanilla investing).  But buying into an asset class is not as dangerous.  The world global equity markets, taken as a whole, are not likely to hit zero [though there are no guarantees].

Let's look at what happened to Siobhan. She's saving £3000 a year into an ISA and had savings of £30,000, split 50:50 between gilts and shares. Held in a FTSE100 tracker and a medium-term gilts etf.

Siobhan has watched her shares drop 40% in value, her  savings are down to £24,000. To rebalance, she should now sell £3000 of her gilts and buy shares. This is tough. She has just watched two-years worth of hard saving disappear in the market falls. It feels like throwing good money after bad. 

So what did Siobhan do? Well if she just obeyed the rules, that would be that. But in the real world very few people do stick with a simple set of proven rules. Otherwise we'd all be vanilla investors.

First off , the 50:50 split turned out to be too scarey for Siobhan; 40:60  might have suited her better. Note that a portfolio should be tailored  to 'suit' your personality and circumstances more than market conditions. But Siobhan is enough of a vanilla investor to realise that buying when the market is low is better than selling. So she resists the temptation to give up on shares completely and sell out. She is saving for the long term, she is not going to change her strategy every time the market drops.

She still can't bear to sell any of her £15,000 of gilts though, which have been rock steady and such a comfort. So she compromises. She decides to channel all her dividends, interest and futures savings into the 'shares' side of her portfolio. She figures it will take her about 18 months for her monthly savings to push her current £9000 of shares back up to £15,000.

In fact the market rebounded. Within a year and [with the new money she was putting in] her shares were back and more - to  £17,000. Time to sell some shares  to rebalance ? No - selling £1000 worth to put into gilts is not economic, because of broker's commission and admin charges. Siobhan continues the strategy that suits her - any 'new' money [monthly savings, interest, dividends] she will now channel into the gilts side.

The rules on when to rebalance

'Every couple of years' sounds a bit vague, but in tests  its been shown to work ok. In theory rebalancing should happen when the portfolio gets out of balance. That's another simple statement with big implications: there is no timescale. Taken literally you could rebalance every day [big funds do this] or, if you wait for a major imbalance to occur years could go by before action was needed.

As a simple rule - if you have to top up a holding by 25% or more then do it.  If your 50:50 split slips to 40:60 then it is time to rebalance. In Siobhan's case suppose her portfolio grows to £40,000, of which £16,000 is in gilts. She should sell £4000 shares and invest in gilts to bring everything back into balance. That's a 25% increase in her gilts holding.

As ever - real life bends these rules a bit. Selling [especially shares] costs money - and should be avoided if possible. Instead, re-direct savings and interest flows. Don't forget, in the accumulation phase of your life [ie while you are working] you will be buying into asset classes over the long term. You will be buying more of everything. Selling is not normal.*

Finally, even the '25% top-up' rule should be ignored if you are only switching small amounts of money - say at the beginning of your savings career.  Steve has £3500 in shares, his plan says he should  have £2500. Is it worth selling £1000 of shares? In practice - no.

How much  is the minimum then? 

How long is a piece of string? But since you ask, and you want something practical and simple, and because all our other rules seem to have a a 2 and a 5  in them how about: £2500. 

As in - if you are buying and selling stuff over the long run - do it in decent-size chunks.  £1 is too little, £5000 is fine, and somewhere in the middle is the minimum you should aim for.

Yeah but why £2500 exactly?

It ain't exact, but its something you should be aware of. If  at 65 you look back over your savings career and your purchases have been in chunks of £500 rather than £5000, then commission and charges will add up. Seriously. If you aim to keep purchase costs under 1%, and you get charged £12.50 a trade, and stamp duty is 0.5% then the minimum size order turns out to be......

£2500

Yup

But what if there is no trading charge or stamp duty?

I remember some Irish based funds didn't pay stamp duty, and brokers offering special deals - effectively zero charges if you bought an Irish-based etf.  But you can't get round the spread.

Spread?

Even with zero charges if you buy something and sell it back immediately you will have lost money. Because of the difference between the buying and selling price. The spread. Vanilla investors resist making trades incessantly. Charges are one reason, the spread is another. The third is... oh, we'll save that for another time. 

Let indolence by your watchword.

*For pensioners [de-cumulators] 'buying'  to rebalance is not normal as you are taking money out of your portfolio to provide an income. Here adjustments are made by changing where these cash flows are coming from - taking more from gilts, or shares; and selling holdings over the long term. Rebalancing by 'buying' is the option you would look at last. Just as rebalancing by  'selling' is what you avoid as a saver. Remember to change your mindset when you retire.