Safe withdrawal rate

[Limey]

Hi, I am new to this forum. I live in England and have spent my 30 year working life working as a professional global bond investor managing money for institutional clients like pension funds and foundations. Now 50 years old I have my accumulated stash and it would appear that I was a moustachian without even knowing it, whilst colleagues bought fancy cars  I paid off debt and invested.

I stumbled across MMM when researching the issue of whether I have enough saved to cut the umbilical cord of full time employment but have a slightly different approach from the SWR and wondered whether I am not alone.

It's very simple, I invest 100% in a very well diversified portfolio of UK stocks. Half the portfolio is in FTS100 "blue chip" big safe and boring stocks with no more than 2% in any one company. The other half is invested across a about 40 stocks picked using the Joel Greenblatt value meodology. These stocks also have to feel right and I don't buy anything with a dividend yield of less than 3.5%. I also do a few other checks such as check the accounts to view earning growth etc... nothing fancy though.

The whole portfolio has a dividend yield of 4.2% and the average dividend growth rate across the fund is higher than this rate. My plan is to spend dividend income only and assume that the ability for my 80 or so companies to generate income will on average rise in line with rising prices (or more!). I buy the stocks directly through an execution only broker with the cheapest dealing costs and I don't pay anyone a cent for advice or management charges.

A key point in all of this is that once the portfolio is invested it really does not matter a jott if the stock market drops like a stone, all that matters is that my companies on average can still pay dividends and that this will gradually rise with inflation. As a pro bond investor my own personal view is that the 60/40... 40/60 (whatever!) mix of bonds and stocks is not all it is cracked up to be when we live in an age of virtually zero real interest rates. If the bond market gets smashed and I can buy Government bonds again at yield of 4+ then I might change my mind but seems to me that until that happens a diversified portfolio of stocks is a better payer of income and a better matcher of inflation.

Curious to hear what others think. Oddly though this does mean my withdrawal rate is close to the 4% rule!

English bloke.

[frugledoc]

Of course it is a good enough strategy but seems like a lot of work and a fair amount of dealing costs to build the portfolio.

I'd rather invest it all in an all world tracker and take the dividends and a bit of capital each year.

[Financial.Velociraptor]

There is an entire blogoverse of investors using a dividend growth strategy.  It works for plenty of people.

[frugal_c]

I think it's a great idea.  Personally I am increasingly leaning towards dividend paying stocks.  Have done some mechanical back-testing on this type of strategy and while the results do not generally beat the index I would feel more comfortable with it.

My only issue has been finding sufficient stocks I am comfortable with.  I would like to switch to an ETF or perhaps mechanical investing but haven't found a good dividend ETF and am still kicking the tires on mechanical investing.

[Interest Compound]

Once you understand that receiving a dividend is mathematically equivalent to selling stock, things will be much easier. And you'll make more money too :)

[UnleashHell]

I don't think its a bad strategy if the dividends are funding your living expenses. If they are exceeding it then you can set aside the surplus for that market dip where dividends stop growing or are cut.

The biggest downside is the constant monitoring of the individual companies for signs of slowing growth and threat to the dividend or share buy backs that they think are the best use of cash. That can be a red flag.

Of course if you quit paid employment and like following this stuff then you can do a lot more research with the time you have - even go around to shareholder meetings! Thats a lot easier to do with UK based companies than it is US ones!!

[boarder42]

Getting paid a dividend is the same as if the stock had gone up and you sold it. Only typically is a lower return. And has different tax implications.

[frugal_c]

Getting paid a dividend is the same as if the stock had gone up and you sold it. Only typically is a lower return. And has different tax implications.

Typically this strategy actually has very similar results to most indexes.  Some studies indicate higher yielding dividend stocks (if you exclude the extremely high yielders which tend to be flaky companies) tend to out-perform the index.

It is not quite the same as selling stock because it mitigates against sequencing issues.   For instance, the PE on US stocks right now is around 20x.   It is entirely possible that this could drop down to 15x or even 10x.   If the market stayed that way for say 10 years you would heavily draw down your portfolio at rock bottom prices.  If you rely on dividends, generally the dividends won't have fallen by 25 or 50% so you would just try to ignore the market value of your portfolio and continue to cash your dividends.   Basically dividends free you from having to worry about market multiples which fluctuate more than dividend yields.

Of course this can go both ways, if the stock market went to very high valuations, say a 30x earning multiple you would probably be better off to sell stock but most FIRE individuals are more concerned with not depleting their hoard than maximizing their returns.   In that case I think a dividend strategy is a good idea.

[boarder42]

so if the market goes down as you say for 10 YEARS.  dividend yielding companies will feel that pain and back off their dividends.  but we dont need to have this arguement.  its been had many times on here.  and historical data favors selling stock to dividends.

[Interest Compound]

Getting paid a dividend is the same as if the stock had gone up and you sold it. Only typically is a lower return. And has different tax implications.

Typically this strategy actually has very similar results to most indexes.  Some studies indicate higher yielding dividend stocks (if you exclude the extremely high yielders which tend to be flaky companies) tend to out-perform the index.

It is not quite the same as selling stock because it mitigates against sequencing issues.   For instance, the PE on US stocks right now is around 20x.   It is entirely possible that this could drop down to 15x or even 10x.   If the market stayed that way for say 10 years you would heavily draw down your portfolio at rock bottom prices.  If you rely on dividends, generally the dividends won't have fallen by 25 or 50% so you would just try to ignore the market value of your portfolio and continue to cash your dividends.   Basically dividends free you from having to worry about market multiples which fluctuate more than dividend yields.

Of course this can go both ways, if the stock market went to very high valuations, say a 30x earning multiple you would probably be better off to sell stock but most FIRE individuals are more concerned with not depleting their hoard than maximizing their returns.   In that case I think a dividend strategy is a good idea.

Incorrect. This is one of the big myths that drives people towards less efficient investing, and earning less money overall (usually at higher risk). Dividends aren't free money. Receiving a dividend is mathematically equivalent to selling stock. Relying on dividend-only withdraws, does not protect against sequence of returns risk.

Yes, I've seen the studies, they always boil down to looking at stocks which paid a dividend consistently over the last 25 years, and seeing if they beat the overall market over that time period. This is classic survivorship bias. Let's look at the general statement:

"Companies that paid dividends in the past, ended up outperforming the market as a whole."

Since dividends are just another way to express returns, this statement is the equivalent of:

"Companies that had consistent returns in the past, ended up with more returns than the market as a whole."

That sounds like a reasonable statement to make.  How is that information actionable?  Shall I then invest money in the stocks which have performed well in the past, hoping they will continue to perform well in the future?  Alarm bells should start ringing on that one.  When you think about it, the whole concept sounds downright silly.  If you look at stocks which continually paid dividends over a long period of time, of course they will have beaten the market.  Because when companies are doing well...they continue to pay their dividends.  But it's the fact that the companies did well which caused them to outperform, not the dividends. And picking those in advance is nothing more than a guessing game.

[frugal_c]

Interest Compound, I think your entire argument is based on the assumption that their is survivorship bias in the studies.  That is not the case.

The sequencing has to do with dividend yields fluctuating less than pe multiples.  There is no survivorship bias there.

[Interest Compound]

Interest Compound, I think your entire argument is based on the assumption that their is survivorship bias in the studies.  That is not the case.

The sequencing has to do with dividend yields fluctuating less than pe multiples.  There is no survivorship bias there.

What does dividend yields fluctuating less than PE multiples have to do with outperforming the index? The claim was that dividend stocks outperform the index. I've yet to find a dividend fund which outperformed the total stock index fund over the last 20+ years. Heck I haven't even found one that outperformed since 2000. Dividend stocks demonstrably, almost as a rule, fall harder during crashes, so I'm not surprised that most dividend funds simply fizzled out.

As boarder42 mentioned, this has been hashed out many times on these forums. This isn't opinion, it's an observational fact. That's simply the way it is. Receiving a dividend is mathematically equivalent to selling stock. No use going in circles over it.

[frugal_c]

Interest Compound,

I agree with you on the premise that paying a dividend is the same as buying stock or investing in a business.  It does sound like a bit of shell game how the capital gets allocated.  I am just trying to counter the argument that higher dividend payers under-perform rather than arguing that they-outperform.   I am just saying their is some evidence that they out-perform so I lower my expectations and assume they should equal-perform.

Here is a an academic study, which would be survivor bias free that higher dividend payers tend to out-perform.   I personally have done my own backtesting and have just found that they tend to equal-perform more or less so take this with a grain of salt.

"We find that high yielding stocks earn positive risk adjusted returns, whereas low yielding stocks earn negative risk adjusted returns. We also detect evidence of non-linearity in the performance of zero-dividend stocks. Controlling for firm size, seasonality and market risk we find a significant positive relation between dividend yields and returns. We conclude that the evidence is inconsistent with a tax-based explanation "

http://www.researchgate.net/publication/222778498_Taxes_Dividend_Yields_and_Returns_in_the_UK_Equity_Market



I think there are 2 arguments, one relates to whether dividends out-perform over long periods and the other relates to whether it is safer in FIRE to have dividend payers than relying on withdrawing capital.  Sequencing is about the second point.   When you are withdrawing your money it doesn't really matter what is better in the long-run, it is important that you survive to see the long-run.
It is nothing magical, it is just that when PE's fall, dividend yields tend to fall less.  They don't move in lock-step.  So with dividend payers you are not dependant on market sentiment EXCEPT for when you initially buy the stock, and in that case you know what the dividend will be.   When you are relying more on selling stock if market sentiment is negative, which can happen even if the economy is doing ok, you can be forced to liquidate at unfavorable and unpredictable prices.

As I said before, it is not magic and it goes both ways.   Sentiment could improve and PE's could be high for the next decade in which case you would be better off selling stock than relying on dividends.   So I think of it that you are minimizing downside with dividends, while you are maximizing upside by selling stock.

There are also tax advantages (in Canada at least) to relying on lower yielding dividends, so if your money is in taxable accounts that needs to be considered.

[Interest Compound]

Interest Compound,

I agree with you on the premise that paying a dividend is the same as buying stock or investing in a business.  It does sound like a bit of shell game how the capital gets allocated.  I am just trying to counter the argument that higher dividend payers under-perform rather than arguing that they-outperform.   I am just saying their is some evidence that they out-perform so I lower my expectations and assume they should equal-perform.

Here is a an academic study, which would be survivor bias free that higher dividend payers tend to out-perform.   I personally have done my own backtesting and have just found that they tend to equal-perform more or less so take this with a grain of salt.

"We find that high yielding stocks earn positive risk adjusted returns, whereas low yielding stocks earn negative risk adjusted returns. We also detect evidence of non-linearity in the performance of zero-dividend stocks. Controlling for firm size, seasonality and market risk we find a significant positive relation between dividend yields and returns. We conclude that the evidence is inconsistent with a tax-based explanation "

http://www.researchgate.net/publication/222778498_Taxes_Dividend_Yields_and_Returns_in_the_UK_Equity_Market



I think there are 2 arguments, one relates to whether dividends out-perform over long periods and the other relates to whether it is safer in FIRE to have dividend payers than relying on withdrawing capital.  Sequencing is about the second point.   When you are withdrawing your money it doesn't really matter what is better in the long-run, it is important that you survive to see the long-run.
It is nothing magical, it is just that when PE's fall, dividend yields tend to fall less.  They don't move in lock-step.  So with dividend payers you are not dependant on market sentiment EXCEPT for when you initially buy the stock, and in that case you know what the dividend will be.   When you are relying more on selling stock if market sentiment is negative, which can happen even if the economy is doing ok, you can be forced to liquidate at unfavorable and unpredictable prices.

As I said before, it is not magic and it goes both ways.   Sentiment could improve and PE's could be high for the next decade in which case you would be better off selling stock than relying on dividends.   So I think of it that you are minimizing downside with dividends, while you are maximizing upside by selling stock.

There are also tax advantages (in Canada at least) to relying on lower yielding dividends, so if your money is in taxable accounts that needs to be considered.

The study looks like survivorship bias to me. But forgetting that for a moment, you acknowledged my point. It seems we both see the same thing, when looking at real-world results there is no outperformance.

Regarding the sequencing question, you still aren't seeing the true picture. If you buy a stock at $100 that pays a $5 dividend, whenever that dividend is paid the stock price permanently drops by $5:

Pre-dividend day:

Stock holding: $100
Cash in hand: $0
Total: $100

Post-dividend day:

Stock holding: $95
Cash in hand: $5
Total: $100

If there's a stock crash and price drops to $50:

Pre-dividend day:

Stock holding: $50
Cash in hand: $0
Total: $50

Post-dividend day:

Stock holding: $45
Cash in hand: $5
Total: $50

Dividends don't matter, you have the same amount of money either way. It doesn't help with sequencing risk.

[frugal_c]

Interest,

There really is a difference if you extrapolate that out.  The stock price post-dividend doesn't matter when you are relying completing on dividends.  It does matter if your investments have a low dividend yield and you need to sell them.  Over the course of a year, if it is a temporary market drop it doesn't matter much but if the market "resets" to a lower PE multiple and you have to sell at those lower rates you can run out of money.   

Again, I keep repeating it but the market multiple oscillates more than the dividend yields (generally speaking) so when you are withdrawing money to live off of you probably want to be drawing from the thing that oscillates less.   If high dividend yield stocks earn the same total return as low dividend yield stocks then why wouldn't you just use dividends as a basis for your income?  Or at the very least let's just agree that people following such a strategy are not fooling themselves.

This is not just me trying to be technical either, the 4% WDR failed miserably in the 70's and early 80's, if you had started in the late 60's.  Just go to cfiresim and check it out.  A big part of this is that the PE of the market worked it's way down and you would have had to sell investments at this lower multiple. 

[Interest Compound]

There really is a difference if you extrapolate that out.

Please show us, so we can point out the error in your math :) If you're willing to take me up on it, and show your math publicly (right or wrong), it will definitely help all the new investors here who don't understand why dividends are always a hot topic here.

I'll kick it off:

Premise, stocks fall 50%, stay there for 5 years, then double.

Year 1 - I buy a stock for $100 with a $5 dividend. My friend buys an otherwise identical stock for $100 with no dividend. We both need to withdrawal $5 a year to live.

Year 2 - Stocks drop 50% and we take our first withdrawal

Me: My holding goes from $100 to $50, then pays a $5 dividend which I take to live on. I'm left with a portfolio worth $45.

Friend: Holdings goes from $100 to $50, then I sell $5 to live on. She's left with a portfolio worth $45.

Year 3

Me: I take my $5 dividend, and am left with a portfolio worth $40.

Friend: She sells $5 worth of stock, and is left with a portfolio worth $40.

Year 4

Me: I take my $5 dividend, and am left with a portfolio worth $35.

Friend: She sells $5 worth of stock, and is left with a portfolio worth $35.

Year 5

Me: I take my $5 dividend, and am left with a portfolio worth $35.

Friend: She sells $5 worth of stock, and is left with a portfolio worth $35.

Year 6

Me: I take my $5 dividend, and am left with a portfolio worth $30.

Friend: She sells $5 worth of stock, and is left with a portfolio worth $30.

Year 7 - Stocks double

Me: My portfolio doubles to $60. I take my $5 dividend and I'm left with a portfolio worth $55.

Friend: Her portfolio doubles to $60. She sells $5 worth of stock, and is left with a portfolio worth $55.

...

Please extrapolate this out in such a way that withdrawing from dividends leaves you with more money, smoother withdrawals...or really anything different from identical performance, and we'll point out where you went wrong.

[frugal_c]

Fair enough, it will take some time put together and I have things to do today.   Maybe later tonight I will post what I am thinking of.

In the meanwhile the problem with your example is that a dividend stock does not go to 0 as it pays out dividends.  Their are earnings to cover dividends so the stock price drop after the dividend is only temporary.   You have also dis-serviced yourself on the non-dividend payer by not assuming growth in earnings.

[frugal_c]

Okay, well I made a simple version.  Let me know if this conveys the idea.

In order to be fair, we should compare two portfolio's with the same net returns.  So I am thinking a portfolio that returns a 4% dividend yield and 0% inflation adjusted growth vs a portfolio with a 0% dividend yield and 4% inflation adjusted growth.  Can we agree that these are the same?

Let's assume you have $100 in both.   Assume a PE of 20 on the growth stock, so it has $5 in earnings.


Here is what would happen if there are no changes to the market multiple and you withdraw funds / receive your dividend 1 year later.

Dividend Portfolio: 1 year later you would get 4$, the stock price would still be $100.

Growth Portfolio: 1 year later, the company has grown earnings by 4% so the $5 earnings is now $5.20.  Assuming a 20 PE still, the stock is worth $104.  You can sell $4 worth and it is not worth $100.  Your share of the earnings after the sale is again $5.

So as long as the PE stays the same they are equal.


Here is what happens if the PE contracts by 50% to a PE of 10 but with no fundamental changes to the economy.  So the growth company still grows earnings 4% and the dividend of 4% is still paid out.

Dividend Portfolio: 1 year later you would get 4$, the stock price would still be $50 but you would own the same amount.  Year after year you would keep getting $4.

Growth Portfolio: 1 year later, the company has grown earnings by 4% so the $5 earnings is now $5.20.  However, with a PE of 10 it is only worth $52.  You sell $4 worth and now have $48 in the stock.  Your share of the earnings after the sale is $4.80.

Growth Portfolio: 2 year later, the company has grown earnings by 4% so the $4.80 earnings is now $4.992.  However, with a PE of 10 it is only worth $49.92.  You sell $4 worth and now have $45.92 in the stock.  Your share of the earnings after the sale is $45.92.

Can you see how if the PE stayed at this level you would eventually run out of money with the growth portfolio?


Now TO BE FAIR, you can flip this and look at what happens if the PE goes up.  With the growth portfolio, you will gradually get a larger and larger share and will definitely beat the dividend portfolio.   That is what I mean when I say the dividend minimizes downside (but does not eliminate it by any means) while non-dividend maximizes upside.

[RetirementInvestingToday]

...
Curious to hear what others think. Oddly though this does mean my withdrawal rate is close to the 4% rule!

English bloke.

Be careful with that 4% Rule.  I'm a UK Investor like your good self and I'm planning to FIRE on a 2.5% (+ expenses of 0.27% currently) SWR.  I'll then adapt if needed as ER progresses.  That's based on this piece of work from Wade Pfau - http://www.retirementinvestingtoday.com/2014/05/further-exploration-of-safe-withdrawal.html. 

[Interest Compound]

...
Curious to hear what others think. Oddly though this does mean my withdrawal rate is close to the 4% rule!

English bloke.

Be careful with that 4% Rule.  I'm a UK Investor like your good self and I'm planning to FIRE on a 2.5% (+ expenses of 0.27% currently) SWR.  I'll then adapt if needed as ER progresses.  That's based on this piece of work from Wade Pfau - http://www.retirementinvestingtoday.com/2014/05/further-exploration-of-safe-withdrawal.html.

If you're really planning to retire on 2.5%, based on someone's ability to predict the future, why invest at all? You'd do just fine with a CD ladder.

[Interest Compound]

In order to be fair, we should compare two portfolio's with the same net returns.  So I am thinking a portfolio that returns a 4% dividend yield and 0% inflation adjusted growth vs a portfolio with a 0% dividend yield and 4% inflation adjusted price growth.  Can we agree that these are the same?

Yes, I agree, if we are talking about price growth. A $5 rise in price is identical to a $5 dividend for a $100 stock. Further in the post you start talking about earnings growth.

[frugal_c]

In order to be fair, we should compare two portfolio's with the same net returns.  So I am thinking a portfolio that returns a 4% dividend yield and 0% inflation adjusted growth vs a portfolio with a 0% dividend yield and 4% inflation adjusted price growth.  Can we agree that these are the same?

Yes, I agree, if we are talking about price growth. A $5 rise in price is identical to a $5 dividend for a $100 stock. Further in the post you start talking about earnings growth.

I AM talking earnings growth.   A company growing earnings 4% should have the same return as a company growing earnings 0% and paying a 4% dividend if you reinvest the dividend.

[Interest Compound]

In order to be fair, we should compare two portfolio's with the same net returns.  So I am thinking a portfolio that returns a 4% dividend yield and 0% inflation adjusted growth vs a portfolio with a 0% dividend yield and 4% inflation adjusted price growth.  Can we agree that these are the same?

Yes, I agree, if we are talking about price growth. A $5 rise in price is identical to a $5 dividend for a $100 stock. Further in the post you start talking about earnings growth.

I AM talking earnings growth.   A company growing earnings 4% should have the same return as a company growing earnings 0% and paying a 4% dividend if you reinvest the dividend.

If you mean earnings, then you're failing to account for the relationship between dividends and the PE ratio. Two otherwise identical companies, one which pays dividends and one which does not, wouldn't have identical PE ratios.

[frugal_c]

I think you are trying to sidetrack the discussion now.  You are just picking through the argument throwing random stuff out.  Can you just look at the example and tell me what is wrong with it rather than trying to change the subject?

If you look through the fist example, you can see the results are the same.  According to EMT, this is what you would expect.  I don't see the issue.

[Interest Compound]

I think you are trying to sidetrack the discussion now.  You are just picking through the argument throwing random stuff out.  Can you just look at the example and tell me what is wrong with it rather than trying to change the subject?

If you look through the fist example, you can see the results are the same.  According to EMT, this is what you would expect.  I don't see the issue.

I'm genuinely not trying to change the subject. The problem with your example is you're assuming there's no relationship between dividends and PE ratio. When you assume an identical PE ratio for the two companies, despite one of them having a dividend, you're implicitly giving them a different valuation. $1 in earnings given away as a dividend affects price much differently then $1 in earnings kept by the company. This directly contradicts the premise that the two companies are otherwise identical.

Because of this, I think it's best to leave PE ratio out of it, as I did in my example.

[frugal_c]

" $1 in earnings given away as a dividend affects price much differently then $1 in earnings kept by the company. "

Do you agree though that they will have the same rate of return if they are both purchased at the same multiple?

If they do give the same rate of return, then the market has to price them the same otherwise the efficient market theory wouldn't be true.

Didn't we start this whole conversation by agreeing that high dividend stocks and low dividend stocks will probably have more or less the same rate of return over long periods of time.   This takes into account the PE ratio you buy them at.  If you can get higher returns from growth stock then efficient market theory is dead.  If you can get better returns from dividend stocks it is also dead.


Let me throw out one more example to really simplify things.    Again assume you have a dividend ETF paying 4%, compare that to an ETF with no yield.    Let's assume you have to withdraw 4% once a year.   In some bizarro, crazy world, if the stock market dropped by 96% but the dividend and earnings power of the companies behind the ETF's were unchanged, what would happen?  With the dividend ETF, you would collect the 4% dividend and survive with the same amount of stock.   With the no yield ETF, you would have to liquidate everything and be wiped out.   Can you not just extrapolate this to lesser price drops and see that when market multiples drop and you have to sell assets you are increasing your draw down rate?

[Interest Compound]

" $1 in earnings given away as a dividend affects price much differently then $1 in earnings kept by the company. "

Do you agree though that they will have the same rate of return if they are both purchased at the same multiple?

If they do give the same rate of return, then the market has to price them the same otherwise the efficient market theory wouldn't be true.

Didn't we start this whole conversation by agreeing that high dividend stocks and low dividend stocks will probably have more or less the same rate of return over long periods of time.   This takes into account the PE ratio you buy them at.  If you can get higher returns from growth stock then efficient market theory is dead.  If you can get better returns from dividend stocks it is also dead.


Let me throw out one more example to really simplify things.    Again assume you have a dividend ETF paying 4%, compare that to an ETF with no yield.    Let's assume you have to withdraw 4% once a year.   In some bizarro, crazy world, if the stock market dropped by 96% but the dividend and earnings power of the companies behind the ETF's were unchanged, what would happen?  With the dividend ETF, you would collect the 4% dividend and survive with the same amount of stock.   With the no yield ETF, you would have to liquidate everything and be wiped out.   Can you not just extrapolate this to lesser price drops and see that when market multiples drop and you have to sell assets you are increasing your draw down rate?

I think I see what you're saying. When selling stock we rely on the market correctly pricing that stock. If the market is being an idiot, dropping a $100 stock down to $4 despite the fundamentals being the same, and I need $4 to live, I'll be wiped out. But in the dividend example, I can ignore all that and just take the dividend.

If the market is being an idiot, couldn't it also drop the dividend stock price to 1 cent post dividend, forcing the company to drop or cut their dividend, also leaving me wiped out? I guess I'm just having trouble imagining a scenario where the market is an idiot, but it doesn't affect both dividend and non-dividend stocks the same way. Indeed, all the evidence I've seen in real-world dividend funds, show they drop harder during downturns. Here's a handy post from Skyrefuge explaining why:

-------------------------------------------------------------
We'll say that at the end of 2015, both Divcorp and Capgains are valued by the market at $1M, and each company has expected earnings of $60k per year (6%). In December 2016, a nuclear weapon explodes in Cleveland. While some knuckleheads cheer ("it's about time!" "I think the fallout actually *improved* things!"), most of the world sees that the population loss and oncoming military action are going to have an enormous effect on consumer behavior, and far fewer people will be buying the awesome fluorescent slap-bracelets that both Divcorp and Capgains make all their money on. Everyone recognizes that the future earnings of the two companies are likely to be cut in half to $30k/year. So those who are unwilling to accept those lower returns sell their holdings. This selling pressure causes the price of both companies to fall dramatically.

Once the price reaches a level where the earnings are sufficient to produce that original 6% return-on-capital, it stops falling, because that's now an acceptable return for investors in the market. But let's see how far they've actually fallen. Both companies made $60k in 2016, and Divcorp paid out its normal $50k in dividends at the end of the year despite the oncoming war. So its market value in January 2017 is $510k ($30k/6% = $500k, plus the $10k in retained earnings). A drop of 49%. But Capgains retained all of its earnings as usual, so its market value in January 2017 is $560k ($30k/6% = $500k, plus the $60k in retained earnings). A drop of only 44%.

So you see that in a "market crash", Divcorp crashes harder.

Of course, this doesn't make Capgains "better" than Divcorp, because Divcorp provided you with retirement income via its dividend. But when you are forced to "eat into principal" of your Capgains holding to fund your retirement, you do so from a position where Capgains has fallen less-far than Divcorp. "Eating into Capgains' principal" at that point only makes your holding shrink to be as small as Divcorp, not smaller. Your eaten-away holdings in both companies will have an equally hard time recovering from those losses; neither is in a stronger position than the other.

I agree that the behavior of stocks during a crash/withdrawal is seldom detailed like this in a dividend discussion, so hopefully this theory, combined with the visual proof that that's how the theory also works in practice, helps you understand why dividends don't protect you against market crashes during the withdrawal phase.

Source: http://forum.mrmoneymustache.com/investor-alley/ive-decided-on-vanguard-but-need-some-help-please/msg584803/#msg584803
-------------------------------------------------------------

[CowboyAndIndian]

...
It's very simple, I invest 100% in a very well diversified portfolio of UK stocks.
...

Limey,

This bothers me. You need an international component for  your allocation.

[Limey]

Enjoying the feedback... always good to hear more than one view... 2 views = a market!

[Playing with Fire UK]

Hey Limey,

It sounds like you've thought about your investment/dividend strategy, it's different to what I'd do, but I'll leave that alone.

When you talk about spending your 4%/4.2% yield, is this the minimum you'd need to have a reasonable lifestyle, or does it include elements that you can cut back on? If there are elements that you could cut back on and still be happy and comfortable, this will protect you from the sequencing risk of poor returns early on your FIRE time. Also have you considered escalating costs in later life for care if required?

Are you happy that the focus on dividends is adequately tax efficient (ISA, pension, £5k tax free allowance etc)? And have you accounted for any tax you'll pay from your annual income? If you are in taxable accounts, then you might want to look at stategic capital sales to mop up the CGT allowances.

If you're happy that you can deal with a market drop, want to stay solely invested in the UK and have taxes sorted then it sounds like all you need to do is check your numbers and hand in your letter!