In this edition of Industry Focus, host Michael Douglass and Fool contributor Matt Frankel discuss share buybacks. They agree that buybacks make sense when a stock is trading for less than its intrinsic value, or to offset dilution caused by employee stock incentives. However, it's not the best move in all cases.

A full transcript follows the video.

This video was recorded on Oct. 9, 2017.

Michael Douglass: Let's start with the initial question, share buybacks and dividends, and let's start with share buybacks specifically. When do share buybacks make sense?

Matt Frankel: There's a few situations where buybacks can make sense. First of all, dividends and buybacks together are kind of ways of a company to admit that they don't need the money as much to grow anymore. In other words, it's kind of a sign of a maturing company. That's why you see companies like Procter & Gamble and Johnson & Johnson pay dividends, but companies that are trying to put all their money into growth, like, say, Netflix, don't.

Douglass: I would even argue that it's a sign, when a company basically says, "We're making more money than we can redeploy into the business itself," that's when buybacks and dividends really make sense. It's essentially saying, "Hey, we're making $1 billion. We can't spend more than $800 million of it effectively for the business, so let's return the rest."

Frankel: Right. Another reason buybacks make good sense is when a stock is trading less than its net asset value, less than its book value, or less than the company thinks it's worth. A good example of that is Berkshire Hathaway, where they set a specific level, I think it's 140% of book, if I'm remembering correctly.

Douglass: It's moved around a little bit over the years.

Frankel: Yeah, it moves over the years, and it's honestly probably going to go up again soon. But where Warren Buffett says, "OK, now it's at a discount to what it's actually worth, so we can use money on share buybacks." So that's another good example of a reason where buybacks would make sense.

Douglass: I'll also throw out there, occasionally in a financial engineering sense, buybacks can make a good sense. Let's just draw a scenario and take an example. A company is paying a 4% dividend, and they can take on debt at 1%. In that case, it would make sense for them to buy back shares, and perhaps even buy back shares taking on some debt so that they could then reduce their dividend exposure, because they're trading -- let's say it's 4%. Let's say it's $4; make it easy -- $4 per share of dividend expense, and trading it for $1 per share of interest expense. And that can basically free up more money that they can then use more effectively.

Now, on the flip side, buybacks don't often make a lot of sense. One of the key things you pointed out, Matt, is that companies can use buybacks effectively when the company is trading for less than it really should be worth. But management is notoriously bad at timing this kind of thing. I would also say, a lot of the times, companies think they're worth a great deal more than they really are. A great example of this is Bank of America. From 2003 to 2007, they bought back around $40 billion in shares. And as a result of doing so, they reduced their share count significantly. But a year and a half later, the financial crisis hit, and the Federal Reserve required them to raise more money. So at a quarter of the price per share, they ended up having to raise about $48 billion. So they had retired 768 million shares, and they then had the issue 3.5 billion new shares. So this was a case where what might have seemed to make sense in the short term really didn't in the long term, and it effectively destroyed shareholder value.

Frankel: It's also interesting to point out that, at the time you're referring to, Bank of America was trading at about double its book value. So the writing was kind of on the wall that it might not be a great time to buy back a ton of shares.

Douglass: Absolutely. The other piece that's key to remember with share buybacks is that management is often incentivized with shares. And to be fair, this isn't necessarily a bad thing. It essentially says, guess what, a significant portion of your compensation is going to be tied to these shares. Drive shareholder value, and you will drive value for yourself as well. So it's intended to align management's goals, financially and personally, with shareholders. But all too often, the share buyback is essentially just papering over that dilution and buying back the shares that management has. So what you really want to see with the share buyback for it to be really effective is for it to actually reduce share count.

Frankel: Definitely. It's usually easy to keep track of a company's share count over time. If you look at a Yahoo! Finance or pretty much any kind of stock quote that's out there, you can usually see the share count at the end of each fiscal year. It's also in the company's annual reports. So that's kind of a good metric to check on.

Douglass: Yeah, absolutely. One more point about share buybacks: One of the things that makes them attractive to companies in a lot of ways is that they aren't as difficult to cut as dividends, whereas dividends, there's sort of a different expectation.

Matthew Frankel owns shares of Bank of America and Berkshire Hathaway (B shares). Michael Douglass owns shares of Berkshire Hathaway (B shares) and Johnson & Johnson. The Motley Fool owns shares of and recommends Berkshire Hathaway (B shares), Johnson & Johnson, and Netflix. The Motley Fool has a disclosure policy.