Interesting chart!
However presumably some of the increased/decreased performance of the “dividend initiator”/“dividend eliminator” categories is not something you could trade on in advance: i.e. you’d have to own a company already to profit from the growth when it announces it will start paying a dividend for the first time.
Probably.
But I assume most dividend-growing companies have increased their dividends in previous years.
True.
The (IMO) major message still holds: over time, dividend paying companies perform better than dividend non-payers.
Edit: “better than” correcting “better that”
According to portfolio visualizer, $100 invested in the US Stock Market from 1973 to 2021 would have grown to $15,182. Is something off in this chart? And why didn’t they include a market cap-weighted investment as comparison?
I wonder if AAPL doesn’t have an outsized impact on the returns? Definitely should be compared with capweighted anyway.
I believe the main point of the chart is to compare subsets (dividend growers, payers against non-payers, etc) of the S&P 500 against each other - as the title says: “returns of S&P 500 index stocks by dividend policy”.
Anyway, I didn’t want to argue with you, I just wanted to provide a data point.
BTW, the chart image is linked to its web source if you’d like to dig deeper.
In what regard?
Inflating the Dividend Growers and Initiators? Flattening the Dividend Non-Payers? Something different?
Apple didn’t pay any dividends until 2012.
Apple presumably moved in August 2012 from the S&P 500 subset of Dividend Non-Payers to the subset of Dividend Growers and Initiators.
To me, though, the slope of all of the subset curves looks unchanged since the dot com bust. 2012 doesn’t stand out.
If you squint a little, the slopes of the subset curves are relative to each other unchanged since about 1980 (though that’s possibly due to the graph resolution for the curves on the left hand side of the chart).
Certainly doesn’t look like there weren’t lots of inflection type cross-overs - maybe one or two between the Dividend Non-Payers and the Dividend Cutters & Eliminators.
Looked at the original link, indeed it can’t be aapl (and all the categories in the graph are equal weighted, not cap weighted which wasn’t obvious to me).
I wonder if the same chart but cap weighted would tell a different story (it’s already interesting that cap weighted S&P already beats the best sub category).
Ah, yes, cap weighted would tell a different story: if you can’t or don’t want to pick stocks, buy the cap weighted S&P 500 (or some other index you like). You’ll do best over long periods of time. Period.*
Going back to the original ask “Only invest in shares that pay no dividends” would be picking stocks, though - i.e. the Dividend Non-Payers (or worse, including the Dividend Cutters & Eliminators) …this passive ETF doesn’t exist for a reason.** And we’ve come full circle. 
* Outperforming or even just matching the index isn’t always the goal, though. Perhaps one wants to maximize tax efficiency, as the originator of this thread, or better than index yearly cash flow / dividends to offset current expenses (as I aim to do). Active investing can help achieve those goals. Sometimes it even beats the index significantly.
** Imagine you want to launch a new passive ETF or mutual fund with “companies that pay no dividend (or have cut or eliminated)”. Do some backtesting and you’ll quickly conclude “no way” (not saying that past performance is indicative of future performance, but backtesting is just on the 101 list for portfolio management - unless you are Cathie Woods, of course).
I have come to a conclusion about dividend oriented portfolios: Not for me. I have also stopped pursuing a Non-Dividend ETF.
If you’re interested more in the Dividend vs. Non-Dividend debate, read on.
The debate usually comes from two points of view (if you want to read an extensive, heated but civilised debates between empiricists and dividend advocates, read this lenghty discussion here https://community.rationalreminder.ca/t/dividend-relevance/13616 - registration required):
On the one hand, there are those who argue empirically that dividends are irrelevant and that one would get better returns when investing in companies that exhibit traits of dividend paying companies but that do not focus on paying the most dividends (Factor investing in Value and Profitability). Also, dividends are less tax-efficient in most jurisdictions.
On the other hand, there are those who argue that without paying dividends, there would have been no profit in the stock market. Or that empirical evidence does not cover some of the relevant practical implications of getting dividends.
I think dividend oriented stocks offer a huge psychological advantage: They are good for those who need a feedback on their investments. They feel better when they get something in their bank account. This keeps them going (when it does not lead them to spending “free money”).
It may be a good thing for some, but not for me. I’d prefer share buybacks way more.
My conclusions about dividends so far:
A note about BRK.B:
[Minor edits for readability and orthography]
This is what sums it up best: it may be good for me, but not for you.
I.e. it depends on your personal situation. You don’t need steady cash flow (from investments), I do.
It may be good to distinguish between
Apple’s dividend, for instance, has never been particularly high. Yet they‘ve steadily held or increased per stock dividends for a decade. And been a great investment during that time - even in stock price, before dividend distributions and reinvestments.
On the flip side, many of the very highest dividend yielding stocks haven’t been that great an investment. At 5% dividend yield, you can be glad if your stock‘s price holds flat.