What would this look like from a practical standpoint, though? Dividends are a taxable event because you're giving someone money. In a buyback, the value of the stock simply goes up, which isn't a taxable event. How do you determine the cost basis on something like that? If there's a stock buyback over the course of 6 months, how do you determine which proportion of the price increase is due to the buyback, as opposed to inflation or normal growth?
In a buyback, a shareholder must sell shares back to the company. When they do, the shareholder's gains are taxed with capital-gains tax.
Furthermore, whenever shareholders sell shares after the buyback, their shares are generally worth more, so they pay increased capital gains, too.
I don't know which yields more tax revenue in the long run, but buybacks definitely generate some tax income.
The only time I could see a "buyback" influencing share price without taxation is if the buyback has a price cap, and every shareholder believes that the company is worth more, so they don't sell any shares back. This is exceedingly unlikely unless the company is very thinly traded.
From my perspective, the key difference with a buyback is the winnowing of shareholders to those who believe more-strongly in the company's future. With a dividend, every shareholder gets a small amount of cash. With a buyback, some shareholders get cash for their shares, and every other shareholder gets a bigger slice of a company that is worth less (total) money.
If the market is efficient (and it isn't, but it can be) then following a buyback, one would expect the market capitalization of the company to be smaller, as the company has paid out money. After 6 months, if the market cap of the company, plus the funds expended in the buyback, is greater than the inflation-adjusted pre-buyout market-cap, then the company has created shareholder value through some other mechanism (even if the mechanism is pure psychology).
>one would expect the market capitalization of the company to be smaller, as the company has paid out money
This doesn't make sense to me. If I spend $X to buy a widget worth $X, I've paid out money, but I haven't net lost anything. I'm worth the same in total. If the widget is a share in a company, even my own, that shouldn't change the fact that the net change in total value is zero.
When a company buys its own shares it’s essentially cancelling them: they become essentially worthless.
Look at if from a different angle: the ones selling the shares to the company are the shareholders. In aggregate, the shareholders own the company valued at $B (market cap) before the buyback, and the company valued at $A plus the cash paid $X after the buyback. If $A=$B they are creating $X out of thin air.
Even simpler to grasp: you have a business 50/50 with a partner. There is $1m in the company account, you reach an agreement and he will sell his interest back to the company for $1mn in cash. You are sole owner now. The company doesn’t have the cash anymore. How much was the company worth before? How much is it worth now?
> After 6 months, if the market cap of the company, plus the funds expended in the buyback, is greater than the inflation-adjusted pre-buyout market-cap, then the company has created shareholder value through some other mechanism
The mechanism may be called profit, earnings, free cash flow... at least for the old-fashioned companies that bring in more money that they spend over the six-months period.
I think it is mostly psychology via ratios. For example a number of institutional investors go by EPS. If shares in this industry are trading at a 20 x multiple and your companies shares are trading at a 10x multiple, you can try to buy back shares to get your multiple higher. Again, just one example but I'm sure the boards of directors, who are elected mainly by the institutional investors determine what will multiples and ratios they need to hit to be in the magic zone for algorithms and also investor psychology.
Buybacks get you a higher EPS number, I don’t see why would you expect the multiple to increase. But it’s true that there are many effects in play: signaling, offer/demand of shares, apparent improvement in ratios vs peers...
> If the market is efficient (and it isn't, but it can be) then following a buyback, one would expect the market capitalization of the company to be smaller, as the company has paid out money.
The company's cash flow is the same, but there are fewer outstanding shares, so earnings per share goes up. Earnings per share matters because when it increases that implies that future dividends per share will also increase.
If investors expect higher future dividends, that means that the stock is worth more, so it goes up.
You tax the buyback itself, right? You tax the money the company is spending on its own stock. That's the only step of the transaction where money is changing hands
Isn’t that a double tax? Taxed at buyback time, and taxed when shareholder cashes out and realizes their gain? Though I guess a small (10%??) tax could still make buyback tax + dividend long term gains is still less than normal income tax.
I'm not trying to be obtuse, I'm genuinely curious how this would work and what changes to tax law would need to be made.
Would you tax the entity buying the stock (in case always a company since it's a buyback; which I think would be the only example of taxing a stock purchase), or the entity selling the stock (either an individual or a company, which is already potentially taxed as capital gains)?
If the goal is to tax it similar to dividends you would tax the person selling the stock but it's already taxed as capital gains, either long term (0%/15%/20%) or short term (marginal bracket rate).
