Dividends vs. drawdowns

Dividends vs. drawdowns
I read this Terry Smith article some time ago (also available on the FT site here), and thought I should finally get around to digging into it a bit more. I think you can question the merits of using price to book as an appropriate metric, particularly with low capital companies who generate cash with few physical assets. I’m sure you could also disappear down a rabbit hole attempting to analyse retained earnings for a bunch of companies, particularly a disparate collection such as those in Fundsmith. But I am most interested in the distinction between using dividends for an income stream versus selling fund units (or one’s shares) as a means to generate income - to quote the article:

"The need to get spending money from your investments once you've retired is obvious. But why does it have to come from dividends?"

If I had an investment that paid out 4%, something like a FTSE 100 tracker, how would taking the 4% dividend compare to selling units to generate the same income? The FTSE 100 hasn't performed particularly well over recent years compared to other indices, whereas in terms of capital growth Fundsmith has been one of the better places to park your cash. It's only a small sample, but comparing these two investments, let's see how Mr Smith's assertion stacks up.

Income dilemma

I’m starting from a position that I need my investments to provide me with some income. There are two ways I can generate this:

• Have an investment that pays out a steady dividend
• Sell a slice of my investments at regular intervals

I’m going to take the FTSE 100 as my initial virtual investment – or more precisely a couple of iShares index funds to compare, one an accumulation fund that automatically reinvests dividends, one a distribution fund that pays out dividends quarterly.

I begin with an imaginary £100000 (just because it's a nice round number). I have wound back the clock to January 2010 when the accumulation fund I’ve chosen was started, and tracked performance forward from there until the end of June 2019. To make life easier I’m ignoring fees.

The capital of the distribution fund will remain untouched, and the dividends are not reinvested. For the accumulation fund I will sell as many units as I need to generate the same income provided by dividends from the distribution fund. I will also sell these units in the same months as the dividends are paid out.

Lets start with some stats showing how the trackers performed before I sell any units.

FTSE 100 accumulation tracker:

• Starting value January 2010: £100000
• End value June 2019: £200829
• Total return % change: 101% increase
• Total return: £100829

FTSE 100 distribution tracker:

• Starting value January 2010: £100000
• End value June 2019: £140647
• Capital % change: 41% increase
• Dividends paid: £43941
• Total difference (capital + dividends): 85% increase
• Total return: £84588

As might be expected, dividends being reinvested automatically into the accumulation fund, and the wonders of compounding provide for a much better performance. The additional 16% that the accumulation fund provides sounds a little abstract, but an additional £16k in the bank certainly isn’t.

But I need an income, so how does the accumulation tracker compare as an investment if I need to sell off quarterly slices in order to pay the bills?

Accumulation tracker:

• Starting value January 2010: £100000
• End value June 2019: £141389
• Capital % change: 41% increase
• Total value of units sold: £43941
• Total difference (capital + dividends): 85% increase
• Total return: £85330

In other words, generating an income through dividends, or taking a slice of the fund gives pretty much the same result. Changing the timing of unit sales would affect the returns, but if this was my income I wouldn’t necessarily be able to delay paying bills to wait for a favourable price at which to sell units.

This also doesn’t take into account fees, for the two funds I’ve chosen the ongoing charges are the same, but depending on the platform in which you choose to invest, transaction fees would nibble away at the small advantage that the accumulation fund offers. There is also the hassle of having to sell the units etc. not a big deal maybe, but compared to simply receiving a quarterly dividend in your bank account, it’s an effort that can be avoided.

So it doesn't look like the assertion from the article above holds much water in this particular example. At the time of writing the FTSE 100 ETF has a yield of 4.4%, and as can be seen here, aside from the financial crash years around 2009, the current yield is at the upper end of it's historic range. Fundsmith Income fund has a much lower yield of around 1.5%. Lets see how the Fundsmith accumulation fund would fare providing the same income as the FTSE 100 tracker, if I had to sell off units of Fundsmith.

Mr Smith vs the index
Fundsmith launched in November 2010, 10 months or so after the launch date of the iShares accumulation fund above, so I will adjust the unit sales to match. Basic stats as above to start with:

Fundsmith accumulation:

• Starting value November 2010: £100000
• End value June 2019: £453020
• Total return % change: 353% increase
• Total return: £353020

Now if I sell units to provide for the same dividend income as the FTSE 100 distribution tracker above we get the following:

• Starting value November 2010: £100000
• End value June 2019: £360616
• Capital % change: 261% increase
• Total value of units sold: £41444
• Total difference (capital + dividends): 302% increase
• Total return: £302060

Even when Fundsmith has to pay out an additional few percentage points in income to match the FTSE 100 ETF the differences are pretty stark. Selling the units would give the same income as the tracker, but the capital is up over 260% compared to the 41% of the tracker. Using this second example, the approach outlined in Mr Smith's article appears convincing.