I think kbg's analogy to a house is a great idea. Let's look at a single-family house, and say that it has a useful life of 50 years. (The IRS says 27.5 but let's say 50.) And we'll leave any land out of it and just look at the house.
Built 1950 - original owner (A) paid $50,000.
1960 - A sells to B for $70,000.
1970 - B sells to C for $90,000.
1980 - C sells to D for $65,000. (The house is starting to get old.)
1990 - D sells to E for $25,000.
2000 - E demolishes the no-longer-useful house.
Investor |
Carrying costs |
Capital gain |
"Dividend" |
Net total gain |
A |
10 years of house maintenance |
$20,000 |
10 years of living |
$20,000 + 10 years of living - 10 years of house maintenance |
B |
10 years of house maintenance |
$20,000 |
10 years of living |
$20,000 + 10 years of living - 10 years of house maintenance |
C |
10 years of house maintenance |
$-25,000 |
10 years of living |
$-25,000 + 10 years of living - 10 years of house maintenance |
D |
10 years of house maintenance |
$-40,000 |
10 years of living |
$-40,000 + 10 years of living - 10 years of house maintenance |
E |
10 years of house maintenance |
$-25,000 |
10 years of living |
$-25,000 + 10 years of living - 10 years of house maintenance |
Totals |
50 years of house maintenance |
$-50,000 (which was the original construction cost) |
50 years of living |
50 years of living - 50 years of house maintenance - construction cost |
Some investors made money, some lost money. The net capital gain is only the inverse of the construction costs. It's really the bottom-right cell that tells the tale: across all the investors, they are ahead by the dividend minus the carrying costs minus the construction cost. So for all investors as a whole, it's the same as if the original one had kept the asset its entire life: he bought 50 years of living in the house for the cost of construction plus 50 years of maintenance. Makes perfect sense. And it jives with HB's advice to not consider your residence to be part of your portfolio: you're buying a place to live.
So the only positive that all investors as a whole got out of the property was the "dividend".
On to stocks. A hypothetical company's founder (A) spends $50,000 to get off the ground, and in 1950 goes public, issuing 10,000 shares at $10/share. Let's say for the sake of simplicity that he doesn't end up with any of the shares: he put in his $50,000, gets $100,000 from the initial offering, and no longer owns the company. Let's say this is a Berkshire-style company that "never" pays a dividend.
The company nets a profit of $25,000 every year, but it never pays dividends. It puts the money in T-bills or a savings account. (We'll ignore interest.)
Investors B and C each buy 5,000 shares at $10/share for a cost of $50,000 each. In 1960, B wants to "make his own dividend" and sells half his shares to D for $35/share. In 1970, C sells half his to E for $60/share. By 1980, there are serious problems with the company, and it's valued at just the amount of money that it has in the bank. (I know the P/E has been crazy all this time, but let's go with it.) E cashes out (by selling to F; every transaction must have a buyer) for $75/share. In 1990, the company winds up operations. We'll deal with different scenarios as to why and how.
For now, consider how much each investor, as well as the sum total of all investors, are doing as time goes by. Note that the bottom-right corner, indicating the sum of all the gains by all the investors, never changes.
1950. Company valued at $100,000. Company has $0 in retained earnings/savings.
Investor |
Shares owned |
Amount spent on stock |
Amount earned from stock |
All-time total gain |
A |
0 |
$50,000 |
$100,000 |
$50,000 |
B |
5,000 |
$50,000 |
$0 |
$-50,000 |
C |
5,000 |
$50,000 |
$0 |
$-50,000 |
Totals |
10,000 |
$150,000 |
$100,000 |
$-50,000 |
1960. Company valued at $350,000. Company has $250,000 in retained earnings/savings.
Investor |
Shares owned |
Amount spent on stock |
Amount earned from stock |
All-time total gain |
A |
0 |
$50,000 |
$100,000 |
$50,000 |
B |
2,500 |
$50,000 |
$87,500 |
$37,500 |
C |
5,000 |
$50,000 |
$0 |
$-50,000 |
D |
2,500 |
$87,500 |
$0 |
$-87,500 |
Totals |
10,000 |
$237,500 |
$187,500 |
$-50,000 |
1970. Company valued at $600,000. Company has $500,000 in retained earnings/savings.
Investor |
Shares owned |
Amount spent on stock |
Amount earned from stock |
All-time total gain |
A |
0 |
$50,000 |
$100,000 |
$50,000 |
B |
2,500 |
$50,000 |
$87,500 |
$37,500 |
C |
2,500 |
$50,000 |
$150,000 |
$100,000 |
D |
2,500 |
$87,500 |
$0 |
$-87,500 |
E |
2,500 |
$150,000 |
$0 |
$-150,000 |
Totals |
10,000 |
$387,500 |
$337,500 |
$-50,000 |
1980. Company valued at $750,000. Company has $750,000 in retained earnings/savings.
Investor |
Shares owned |
Amount spent on stock |
Amount earned from stock |
All-time total gain |
A |
0 |
$50,000 |
$100,000 |
$50,000 |
B |
2,500 |
$50,000 |
$87,500 |
$37,500 |
C |
2,500 |
$50,000 |
$150,000 |
$100,000 |
D |
2,500 |
$87,500 |
$0 |
$-87,500 |
E |
0 |
$150,000 |
$187,500 |
$37,500 |
F |
2,500 |
$187,500 |
$0 |
$-187,500 |
Totals |
10,000 |
$575,000 |
$525,000 |
$-50,000 |
1990.
Scenario 1: The company craters. It burns through all its cash reserves and goes under. The final numbers are the same as 1980.
Scenario 2: The company has $1,000,000 in the bank. The company smoothly ceases operations and distributes its retained earnings as a dividend. The final numbers are:
Investor |
Shares owned |
Amount spent on stock |
Amount earned from stock sale |
Dividend |
All-time total gain |
A |
0 |
$50,000 |
$100,000 |
0 |
$50,000 |
B |
2,500 |
$50,000 |
$87,500 |
$250,000 |
$287,500 |
C |
2,500 |
$50,000 |
$150,000 |
$250,000 |
$350,000 |
D |
2,500 |
$87,500 |
$0 |
$250,000 |
$162,500 |
E |
0 |
$150,000 |
$187,500 |
$0 |
$37,500 |
F |
2,500 |
$187,500 |
$0 |
$250,000 |
$62,500 |
Totals |
10,000 |
$575,000 |
$525,000 |
$1,000,000 |
$950,000 |
Just like the house example, in both the "collapse" and "smoothly cease" scenarios, the net gain for all the investors as a whole is the amount of the dividend minus the initial startup cost. And it was the potential of the dividend that kept up any interest in owning the stock. Ultimately, every company will have a day when it stops operating, and this is the final destiny of all of them.
This ability to distribute the company's assets to shareholders is what gives value to any company. If the company were truly forbidden to distribute dividends, then you end up with Scenario 1, except much worse because such a stock would never have had any value.
I think it's clear that without dividends, the various investors are passing the same shares around, with no effect on the total net gain, which remained at the inverse of the startup costs throughout. That is zero sum.
I also did the math on an alternative scenario where the company builds a reserve of $100,000, then pays out its $25,000 every year as a dividend.
The 1950 initial condition is the same.
1960. Company valued at $150,000, $100,000 in the bank. B's shares are worth $15/share.
Investor |
Shares owned |
Amount spent on stock |
Amount earned from stock sale |
Dividend |
All-time total gain |
A |
0 |
$50,000 |
$100,000 |
0 |
$50,000 |
B |
2,500 |
$50,000 |
$37,500 |
$75,000 |
$62,500 |
C |
5,000 |
$50,000 |
$0 |
$75,000 |
$25,000 |
D |
2,500 |
$37,500 |
$0 |
$0 |
$-37,500 |
Totals |
10,000 |
$187,500 |
$137,500 |
$150,000 |
$100,000 |
1970. Company valued at $150,000, $100,000 in the bank. C's shares are worth $15/share.
Investor |
Shares owned |
Amount spent on stock |
Amount earned from stock sale |
Dividend |
All-time total gain |
A |
0 |
$50,000 |
$100,000 |
0 |
$50,000 |
B |
2,500 |
$50,000 |
$37,500 |
$137,500 |
$125,000 |
C |
2,500 |
$50,000 |
$37,500 |
$200,000 |
$187,500 |
D |
2,500 |
$37,500 |
$0 |
$62,500 |
$25,000 |
E |
2,500 |
$37,500 |
$0 |
$0 |
$-37,500 |
Totals |
10,000 |
$225,000 |
$175,000 |
$400,000 |
$350,000 |
1980. Company valued at the $100,000 it has in the bank. E's shares are worth $10/share.
Investor |
Shares owned |
Amount spent on stock |
Amount earned from stock sale |
Dividend |
All-time total gain |
A |
0 |
$50,000 |
$100,000 |
0 |
$50,000 |
B |
2,500 |
$50,000 |
$37,500 |
$200,000 |
$187,500 |
C |
2,500 |
$50,000 |
$37,500 |
$262,500 |
$250,000 |
D |
2,500 |
$37,500 |
$0 |
$125,000 |
$87,500 |
E |
0 |
$37,500 |
$25,000 |
$62,500 |
$50,000 |
F |
2,500 |
$25,000 |
$0 |
$0 |
$-25,000 |
Totals |
10,000 |
$250,000 |
$200,000 |
$650,000 |
$600,000 |
1990.
Scenario 1: The company craters. It burns through all its cash reserves and goes under. The final numbers are the same as 1980.
Scenario 2: The company has $100,000 in the bank. The company smoothly ceases operations and distributes its retained earnings as a dividend, in addition to its normal dividend. The final numbers are:
Investor |
Shares owned |
Amount spent on stock |
Amount earned from stock sale |
Dividend |
All-time total gain |
A |
0 |
$50,000 |
$100,000 |
0 |
$50,000 |
B |
2,500 |
$50,000 |
$37,500 |
$287,500 |
$275,000 |
C |
2,500 |
$50,000 |
$37,500 |
$350,000 |
$337,500 |
D |
2,500 |
$37,500 |
$0 |
$212,500 |
$175,000 |
E |
0 |
$37,500 |
$25,000 |
$62,500 |
$50,000 |
F |
2,500 |
$25,000 |
$0 |
$87,500 |
$62,500 |
Totals |
10,000 |
$250,000 |
$200,000 |
$1,000,000 |
$950,000 |
For Scenario 1, the situation is obviously much better for investors with dividends paid out as they're earned. For Scenario 2, the total amount earned by investors is identical to the non-dividend company.