It doesn't matter if the stock is undervalued (which is a strictly subjective premise anyway - undervalued by what measure?). If it's overvalued that just reduces the value proposition of the buyback. The value proposition is always positive no matter what price is paid (that however says nothing about the specific value proposition, merely that it's a positive number).

I'd argue that GM is closer to undervalued than anything else. It's capable of yielding at least $6 billion in net income over the next four quarters, giving it in the neighborhood of a 10 multiple. The economy looks to remain strong'ish over 12 months, the consumer is spending and able to borrow at low rates, which will continue to float GM's sales.

It's also worth noting it's not enough to only take the over / under valuation of the stock into account when doing a buyback. You have to account for use of cash - that is, what's the best use of your cash holdings. If there are no great alternatives - for any number of reasons - there is nothing wrong with doing stock buybacks when the stock price is richly valued (especially if the cash held is producing mediocre returns, as cash does today). Buying the stock back may be a vastly superior choice to sitting on the cash when it comes to producing returns for shareholders.

Chipotle is famous for doing this despite their very hefty valuation and relatively rapid expansion. Has turned out very well for shareholders.

"It doesn't matter if the stock is undervalued."

Warren Buffett has written the opposite of this numerous times. Paying $1 for something that is worth 80 cents (even if it's your own stock) destroys shareholder value. The math and logic behind this is so basic, I don't understand how anyone could believe otherwise.

It doesn't matter if the stock is undervalued for this narrow question of dividends vs stock repurchases.

If an outside investor buys stock, of course, under/overvaluation matters.

Sometimes it's easiest to illustrate the flaw in a theory by using extreme examples, so please forgive me for the hyperbole.

Let's say you have a company with only $100M in discounted future earnings (and no significant balance sheet assets), but it is trading on the market for $1B. This 10x intrinsic value scenario isn't common in real life but we definitely see examples like it during market bubbles.

The company wants to distribute $1M back to shareholders. If it pays a dividend, shareholders receive the full million less taxes - so perhaps $600-750k in their pockets.

If on the other hand the company buys back $1M in stock, they are buying 0.1% of $100M in future earnings, or $100k.

If my post doesn't make sense, google "Buffett on buybacks." He's written about the subject in a number of his annual letters.

As adventured mentions, it doesn't matter what the stock price is.

A stock repurchase is a way to move money from the company to the stockholders. Exactly the same as dividends.

In an efficient market, the effects on market capitalization (outstanding stock * stock price) of dividends and stock repurchases will be exactly the same.

Yeah, my mistake. I guess my argument was against how there was "nothing scary about them." This article wouldn't exist if there wasn't some concern about buybacks.

People have concerns over all kinds of innocuous things, like eg shorting a stock, or GM food, monads, or modern nuclear power stations.

Can you explain that?