Ex-dividend purchases?

[eepee]

If you are playing a dividend game i.e. buying shares for the dividend, then moving on to another one, surely there is an advantage to buy them after the ex-dividend date?

Ideally you buy a share for the dividend in the expectation that it recovers the dividend cost in a reasonable time.

So buying pre ex-divi date, you are expecting recovery by the total dividend amount. If on ex-divi date the price drops by more than the dividend amount the range of recovery needed is greater.

However if you purchase on ex-dividend date, the range remains the dividend total regardless of price.

Does this anaysis make sense???

Regards
ep

[Alaric]

Does this anaysis make sense???

It doesn't make much sense for a private investor when you consider total return as what you gain in dividend you loose by having to pay a higher purchase price. Some institutional managers might do this when they have internal or external rules to satisfy regarding levels of dividends.

For private investors dividends are on balance taxed slightly higher than capital gains.

[UncleEbenezer]

For private investors dividends are on balance taxed slightly higher than capital gains.

True.

One more tiny advantage is that buying ex-div, you're paying very marginally less in stamp duty for the same net purchase. That applies regardless of your circumstances that affect tax on dividends and/or gains.

[GoSeigen]

If you are playing a dividend game i.e. buying shares for the dividend, then moving on to another one, surely there is an advantage to buy them after the ex-dividend date?

Ideally you buy a share for the dividend in the expectation that it recovers the dividend cost in a reasonable time.

So buying pre ex-divi date, you are expecting recovery by the total dividend amount. If on ex-divi date the price drops by more than the dividend amount the range of recovery needed is greater.

However if you purchase on ex-dividend date, the range remains the dividend total regardless of price.

Does this anaysis make sense???

ep, there is no significant advantage to buying a security before or after ex date. Investors price shares on the basis that they will either receive or not receive the dividend. There is nothing to gain.

If you are skeptical, think about it this way: you value a certain company's business the day before xd day at price x. Knowing that if you purchase it you will receive a dividend d in a few days' time, you are happy to pay x+d for the share. Let's say you forget to place the trade so you buy it the next morning instead. Nothing has changed overnight about your assessment of the company's value, but now you know you will not receive the dividend. You are therefore prepared to pay only x for the share [for geeks: with a tiny adjustment in line with short money rates].

This is the way every intelligent investor views the payment of dividends so, AEBE, it follows you will gain nothing by choosing to buy either before or after xd.

As others have noted there may be marginal differences for an individual because of of taxation but this is likely to be in the region of a rounding error in the big scheme of things. And it is not to do with the pricing of the share in the market, but depends on your individual tax circumstances.


GS

[richfool]

Good point though re slightly lower stamp duty when buying ex div.

[eepee]

Many thanks for the responses to date. However I feel that I did not make my requirements clear enough.

If a share is priced a day or two before the ex-dividend date at X and the due dividend is D one assumes a drop in price of D. However often the drop in price is not just D. There is an element of sentiment that makes the price change by a greater or smaller amount than D. I'll call this S.

So buying pre ex-div will cost one X and thus the required recovery is D + S (or D - S for positive sentiment).

To buy ex-div means the cost is X - D ± S so the recovery is always just D.

If S is generally negative, i.e. the price drop is D + S. So buying apres ex-div is an advantage. As UncleEbenezer points out there is also the tiny stamp duty advantage since the outlay is smaller.

It is this that I wish to study. If my calculations are correct I want to examine whether price drops greater than D happen more often and by how much. I am just trying to decide whether it is worth writing a short routine to investigate this further over a few months.

Regards,
ep

[Itsallaguess]

If S is generally negative, i.e. the price drop is D [Dividend] + S [Something else]. So buying apres ex-div is an advantage. As UncleEbenezer points out there is also the tiny stamp duty advantage since the outlay is smaller.

It is this that I wish to study. If my calculations are correct I want to examine whether price drops greater than D happen more often and by how much. I am just trying to decide whether it is worth writing a short routine to investigate this further over a few months.

How do you know if S is company-specific or just generally market-related?

If we acknowledge that, all things being equal, buying cum-divi or ex-divi is neutral in terms of company valuation, then if S is market-sentiment related then S is likely to affect many more shares than just the one you're looking at in relation to it being cum-divi or ex-divi.

I think you're in danger of mixing up two completely separate things that can move a share-price around the dividend-date, and perhaps trying to formulate a strategy that's more related to market-sentiment than company-specific price-action related to that particular company, and whether it's paid a dividend yet, or not....

Cheers,

Itsallaguess

[AJC5001]

Many thanks for the responses to date. However I feel that I did not make my requirements clear enough.

If a share is priced a day or two before the ex-dividend date at X and the due dividend is D one assumes a drop in price of D. However often the drop in price is not just D. There is an element of sentiment that makes the price change by a greater or smaller amount than D. I'll call this S.

So buying pre ex-div will cost one X and thus the required recovery is D + S (or D - S for positive sentiment).

To buy ex-div means the cost is X - D ± S so the recovery is always just D.

If S is generally negative, i.e. the price drop is D + S. So buying apres ex-div is an advantage. As UncleEbenezer points out there is also the tiny stamp duty advantage since the outlay is smaller.

It is this that I wish to study. If my calculations are correct I want to examine whether price drops greater than D happen more often and by how much. I am just trying to decide whether it is worth writing a short routine to investigate this further over a few months.

Regards,
ep

Persimmon https://uk.webfg.com/equity/Persimmon goes ex-dividend next Thursday (8th March). Perhaps you could use this as a sample?

Adrian

[GoSeigen]

Many thanks for the responses to date. However I feel that I did not make my requirements clear enough.

If a share is priced a day or two before the ex-dividend date at X and the due dividend is D <i>one assumes a drop in price of D.</i>

I think your logic failed here, right at the beginning. I don't think anyone assumes the price drops by D. That would only be the case if nothing else changes. Of course other things change. On average that means the price will be more expensive if you buy a day later (because GDP grows). In individual cases there will be considerable variation but if you can guess how the price is going to move you may as well buy any share you can predict at any time. And good luck to you.


GS

[eepee]

Since there has been no critique on the mechanism I was curious about, I decided to write a routine that I can use over a few weeks/months.

Most of it was written over the weekend and I should be able to 'go live' on the 7th/8th. There are about 40 shares going ex-div and the program only caters for 28 so I will trim them down (yes, I will keep Persimmon).

Continuations to the thread have made me think further about the 'model'. At this stage it does not matter whether variations in S are related to market-sentiment or company-specific. It is purely the amount of variation that is currently of interest. Once identified as a bias in one direction, a future testing stage could consider the type of sentiment to see if it can be tied into the model for a more confident prediction/selection.

The last post prior to this one does seem to agree that generally the price change is not just by the dividend amount. Well, just as the dividend is known beforehand, so will the extra variation be known on the ex-div date. It would be factored in into the purchase price.

One could even decide that consideration is only given to shares whose price on ex-div date drops by more than just the dividend.

Regards,
ep

[BusyBumbleBee]

Well, eepee, I shall be interested to see the outcome.

A couple of questions and then a comment

What is the price on ex-div day? Opening Offer or closing price or ... ?

Are you taking into account the spread? Ok: this is small for larger companies but for smaller and lower priced shares a spread of say 5 pence has a disproportionate effect on any lower price ex-div.

Persimmon has a large dividend (125p) coming announced on 27th Feb since when the share price has risen. Some people - including me for example - buy shares when a large dividend is announced in the hope/expectation that there will be others (perhaps institutions) who buy for the dividend and that I shall be able to sell it at a profit by selling just before or just after ex-div - and by 'just after' I mean when the market opens. Often the SP fades after the share goes ex-div but with Persimmon there are other factors at play - not least the government diving i with threats etc.

At time of writing the Persimmon shares have ticked up 18 pence. People buying for the dividend?

[Alaric]

Some people - including me for example - buy shares when a large dividend is announced in the hope/expectation that there will be others (perhaps institutions) who buy for the dividend and that I shall be able to sell it at a profit by selling just before or just after ex-div - and by 'just after' I mean when the market opens.

I think it the case that some Fund Managers will be constrained by their internal accounting rules regarding not distributing from capital that they will buy income to boost their fund's apparent performance on distributions. Whether that's anything an alert investor can exploit in all the other factors affecting prices is another matter.

[GoSeigen]

Since there has been no critique on the mechanism I was curious about, I decided to write a routine that I can use over a few weeks/months.

Okay, since you basically ignored everyone, and this is an arbitrage play, how do you intend to hedge your exposure? Will you build the hedge cost into your model?

Or failing a hedge, how do you carry out risk management?


GS

[mickeypops]

Wouldn't the trading fees and stamp duties incurred in such a high turnover strategy eradicate any potential small gains, if they exist?

[kempiejon]

I'm sure there was a HYPer perhaps from TMF not TLF days who ran an experiment and was buying HYP picks cumDiv with a mechanism to sell when XDiv if they reached a limit to include stamp and fees etc and harvest the dividend.

[BusyBumbleBee]

Persimmon ticked up another 19 pence at time of post - people still buying for the dividend tomorrow?

[BusyBumbleBee]

When I first looked this morning Persimmon was down only 111 pence against a dividend of 125 pence. Currently at -108.

[GoSeigen]

When I first looked this morning Persimmon was down only 111 pence against a dividend of 125 pence. Currently at -108.

So what?

Just tracking prices of securities is contrary to the spirit of this site. Can any useful conclusion be added to this reposting of market data (preferably with confidence intervals!!)?



GS

[BusyBumbleBee]

When I first looked this morning Persimmon was down only 111 pence against a dividend of 125 pence. Currently at -108.

So what?
Just tracking prices of securities is contrary to the spirit of this site. Can any useful conclusion be added to this reposting of market data (preferably with confidence intervals!!)? GS

I agree - but the OP had asked the question as to whether to buy before or after the ex-div date and was going to write a routine to track what happens with a selection of shares - including Persimmon. I raised the question as to what he would use as the ex-div price - opening or closing etc and am merely giving him the opening price on the day the share went ex-div.

I often buy a share before a large dividend and sell when it goes ex-div - as explained above. It is one of my investment strategies. Sadly missed out on this one though as I do not know enough about Persimmon.

[eepee]

Oh dear! There was a share called SCAM ?!?!?

Here then a summary of the first live run.

The 40 due shares were picked up from the dividenddata site. For program use these were reduced to 28 (no particular reason just the number of lines I have allocated to the program).
First none-sterling shares were eliminated. Didn't want to have to program conversion rates at this point. In any case the better known shares of this type, such as RDSB and BP, are more likely to be long-term HYP investments. Then got rid of some with the word 'fund' or 'trust'.

I wanted to know if timing made a difference so prices were grabbed at 1000, 1200, 1400 and 1600. It does but not by much. It would however appear that the best time to commit would be around midday when prices seem to be most depressed. This is also true if you are buying for the dividend pre ex-div date - with depressed prices you might get more shares for your allocated amount.

So the result:-

The commission for the calculations is £12.50
The Offer prices were grabbed at 12 from ADVFN on the 7th.
The Offer prices were again grabbed at 12 from ADVFN on the 8th

Out of the 28 shares, 8 reduced in price by more than the dividend.
If instead of buying at minimum cost you had invested the full ex-div fund in shares you could have bought 7 at a greater or similar quantity.

Unfortunately Persimmon, as mentioned by others, did tend to rise during the day so did not qualify for consideration. Interestingly, half an hour early it would have because it dipped at about 11:30. I haven't done a thorough analysis but I don't think it was the high price. Perhaps best to eliminate those shares that are considered safe candidates for HYP.

I will be away for a long weekend so won't be able to contribute to the thread till Monday/Tuesday. I have thus put a screen shot of the run online:-
https://s10.postimg.org/yjlmfkzvd/Erio_EX.jpg

Ouch! I have just posted the image without checking - last time I did this was on TMF - and someone slapped my wrists because apparently there were some 'naughty' adverts on the page. Someone else did point out which link to use. Hope I remembered correctly which one it was!

Regards
ep