Why companies are raising their dividends

By Felix Salmon
December 18, 2013
Matt Yglesias presents the case against dividends today -- and it's a case I'm sympathetic to.

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Matt Yglesias presents the case against dividends today — and it’s a case I’m sympathetic to. But before you can determine whether stocks should be paying dividends, it’s important to understand why stocks are paying these dividends. And the answer is in the chart above.

The blue line, in this chart, US after-tax corporate profits as a share of GDP — and it shows that they’re at an all-time high of around 11%, when they’re normally closer to 6%. This is the chart which should worry anybody invested in the stock market: while the market’s price-to-earnings ratio still seems pretty sane, that’s only because corporate earnings are much higher than they’ve ever been in the past. If this number starts reverting back towards its historical mean, then stock prices are certain to fall, possibly quite sharply.

What investors are looking for, then, is reassurance that the impressive profits they’re seeing today are here to stay, rather than being some kind of historical anomaly. And so that’s also the message that CEOs are seeking to send to their shareholders.

It’s here that dividends start being a lot more attractive than stock buybacks. It’s exactly the same reason that you’d much rather get a thousand-dollar raise than a thousand-dollar bonus. Dividends aren’t bond coupons: they can go down, if they have to — and, in hard times, you can be sure that they will go down. But in general, no company will set a dividend this year which it doesn’t think it can meet next year, and the year after that, and the year after that. A dividend is a company telling the market that the cash it’s throwing off today isn’t some kind of exceptional good fortune, but is rather something that shareholders should get used to, year in and year out for as far as the eye can see.

And that is a message which is much more supportive of a stock price than any stock buyback. (Especially since buybacks are easy to announce, and very few people bother to check whether the companies which announced them actually followed through on their promise.)

If you’re going to return a certain amount of cash to shareholders, then there are lots of reasons why it makes sense to do that with a buyback rather than through a dividend payment. But once a buyback is over, it’s over. A dividend is much more predictable than a buyback program; what’s more, it’s often something which grows predictably, as well. (AT&T has been raising its dividends every year for 30 years.) In a low interest rate environment, a permanently-increasing income stream, even if it only increases in line with inflation, is worth a small fortune.

When a company earns profits, there are lots of things it can do with the money. It can hold on to the profits as a cash balance; it can spend that cash buying back its own stock; it can pay that cash out as a dividend; it can give its employees raises, or bonuses; it can reinvest the money in R&D or other capital expenditure; it can acquire other companies; and so on and so forth. But if you invest your money in employees or capex instead of using it for dividends or buybacks, then that reduces your profits — which in general is bad for your share price.

There are exceptions to the rule, or course — Amazon is a great example of a company with a stratospheric share price, and p/e ratio, despite (or because of) its lack of visible profits. But then again, it was none other than Matt Yglesias who described Amazon as “a charitable organization being run by elements of the investment community for the benefit of consumers”. He can’t really have it both ways. And in any event, if you want to keep your profits high, and send a message to the market that they’re going to stay that way, then it makes a certain amount of sense to boost your dividend.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/

Please stop regurgitating the unprovable theory that buybacks return cash to shareholders. That would only be true if there were no public markets for trading shares, which of course there are. Anyone who wants to get cash back from the company they own shares of is able to do so at any time. Share buybacks do not reward the shareholder, they only sometimes encourage shareholders to bail.

The fallacy that buying shares causes a rise in share price is only true (sometimes) on a transient basis. This is based on the belief that a company is worth its share price x number of shares outstanding, which is rarely true, as usually only a small percentage of shares in most companies are traded during any given short term window. That market cap is a potential valuation of the company, not a real valuation, for not all owners would sell at that price, and there probably aren’t enough buyers at that price, or any price higher.

Once a buyback program is completed, demand for the shares will decrease, and the market will “discover” the new price, and it won’t matter that X shares were bought by the company. The price will still be based on lots of variables, of which EPS is only one.

Yglesias also says dividends are bad for the economy, but this is nonsense, it’s the complete opposite. Unless a company has use for its profits, not distributing those profits effectively takes that money out of circulation. A company like Apple, which took in $170 billion in revenue, also had $64B in gross profit. That means without dividends, it removed $64B from the economy (I know in a perfect, Mayberry pre-1990s world, those accumulated profits would have been deposited in bank accounts where it was lent back to other companies, but that doesn’t happen any more) that would otherwise had been used in downstream trades of goods and services. And it is the trades of good and services that comprise the economy, not the trades of stocks and other assets that may or may not appreciate.

Companies that earn huge profits but do not distribute them to shareholders do their shareholders and the economy a great disservice. The more that companies hoard their profits, the less money is left to circulate in the economy. Buying shares back is just another trade of assets, not a distribution of profits.

Posted by KenG_CA | Report as abusive

“This is the chart which should worry anybody invested in the stock market: while the market’s price-to-earnings ratio still seems pretty sane, that’s only because corporate earnings are much higher than they’ve ever been in the past. ”

I would be worried if the profit margins were higher in the past, but not relative to GDP. That’s irrelevant! The reason they are higher relevant to GDP is that US corps own a lot more factories overseas and sell a lot more goods overseas than they used to. It’s a global market, and some of the profits do make it back home, if not the rest of the economic activity that is part of one semi-useful gross measure of the economy.

Weak story.

Posted by mattmc | Report as abusive

Would using GNP instead of GDP answer mattmc’s point and if so does this prove /disprove his point?

Posted by behindthecurve | Report as abusive

Corporate earnings will be pressured if inflation picks up, but we’re not seeing much hint of that yet. Costs are under control and commodities are slumping.

Share buybacks make sense for a P/E under 16, but after the big rally last year you aren’t seeing much of that. Earnings simply did not rise 20%+ last year for most of the market. So corporations are leaning more towards the dividend — and yes, it is more attractive than a share buyback that may merely be offsetting options grants.

Posted by TFF | Report as abusive

@KenG, I disagree. Share buybacks, when not sanitized by options grants, can have a significant impact on per-share earnings and performance. IBM went through its entire run-up without its top line moving by more than a few percent. Instead of pushing assets into low-return businesses, it bought back shares (often at a bargain P/E) and focused on higher-margin operations. Thus a flat top line, a rising bottom line, and a rapidly rising per-share income. Not sure how well that model will work for them going forward, but they got five years of “growth” out of it.

Posted by TFF | Report as abusive

Even if we want corporate America to invest more of its money, it’s not clear that the best investment opportunities are within the companies spitting off the most cash. Some combination of incentives for (existing or new) banks to raise equity would presumably make sense from a “macroprudential” (if we’re still using that word) standpoint; tighten the screws a bit on required equity ratios and maybe find a way to privilege new equity somewhat over old equity so that banks get to those ratios by raising equity rather than reducing lending. (Indeed, I think part of the reason profits are taking so long to feed into broader growth is that the financial sector is still not fully functional.)

Posted by dWj | Report as abusive

The buffer can almost completely be attributed to lower interest expense and coincidentally lower taxes. After rates start moving up you should see that buffer waste away. As long as we have these low rates you can justify a higher multiple, but as rates increase it should shrink.

Posted by Nevebor | Report as abusive

TFF, yes share buybacks CAN have a significant impact on per-share earnings, but not “will always”, nor “usually will”. There are many factors, including the size of the buyback relative to how many shares are typically traded. It doesn’t increase the value of the company, only the share price, and it doesn’t reward shareholders who want to hold onto the stock. It rewards traders, and it’s just a gimmick to manipulate the share price. It shifts the whole reason for buying shares of a company (or starting any company that might be publicly traded) from wanting to own a business that is profitable to wanting to trade shares, without really caring what the business does.

Besides, if all companies adopted the model of share buybacks instead of dividends, what would a guy like you, who likes dividend paying stocks, do? You would have to transition into a mass psychologist who knew how the traders will react to every bit of news.

Posted by KenG_CA | Report as abusive

@KenG, if I believe a company is undervalued, I’m fine with a share buyback. (I’m less happy with options grants to executives.) If I believe a company is fairly valued, then I would rather have the cash returned via dividend.

Your second paragraph makes no sense. I care more about cash flow and how it is used than about the dividend per se. If a company has a demonstrated history of using part of its cash flow on share repurchases to shrink the share count, I factor that into my analysis. If it has a demonstrated history of using cash flow for internal growth, I factor that in as well. And some companies need to spend a portion of their cash flow on acquisitions as well, even if it is hard to see the immediate benefit of that money. Dividends are nice, but buying the largest dividends you can find is typically a poor strategy.

Note that if a company is actively repurchasing shares, you can in theory sell a portion periodically and achieve much the same effect as a dividend.

Aren’t buybacks almost always a negligible fraction of the transaction volume? Their primary effect is to decrease the denominator in the per-share reporting, thus creating “artificial” growth for a business that is otherwise stagnant.

Posted by TFF | Report as abusive

“If this number [corporate profits as a share of GDP] starts reverting back towards its historical mean, then stock prices are certain to fall, possibly quite sharply.”

Not quite: you’re forgetting the denominator in profits/GDP, Felix. The ratio can mean-revert if GDP rises, too (which is perhaps becoming a less-laughable expectation!) – and in that case P/E ratios remain stable ceteris paribus.

Posted by 00Alex | Report as abusive

TFF, I thought you mostly bought stocks that paid dividends. If they don’t pay dividends, then the only criteria is do you think other people will want to buy the stock at a higher price. As Apple and Amazon have proven, it doesn’t matter if their profits grow or are non-existent, what matters is if people think the company will be more valuable.

If all stocks regularly just repurchase shares, and never pay dividends, then ultimately that action will be priced into all prices, and selling one buyback-adjusted stock to buy another one will result in no real gain.

If a company is actively repurchasing shares, you can sell a portion periodically, just as you can do if they are not repurchasing shares. They don’t make it any easier.

Companies that buy back shares are basically stock traders, and are subject to the same timing issues that other traders have to deal with. The managers are rarely expert at trading shares; sure, they know about the health of the company, and buying back shares sends a strong signal, but they’re not experts on reading the market.

If buybacks are designed to decrease the denominator, as you say, and not to reduce the supply of shares available for sale, then they are even more exposed to external factors that affect their stock price. P/E multiples do not follow a formula, and decreasing the denominator by 2% is no guarantee the share price will increase by 2%. Speculator psychology will have more impact on share price than the buyback, and the shareholders may end up with nothing to show for it.

Posted by KenG_CA | Report as abusive

you say, “Dividends are nice, but buying the largest dividends you can find is typically a poor strategy.”

Which is a strawman.

buying companies who have large dividends and GROW them at a consistent and large rate…now that is a good strategy, when shares are purchased at sound valuations

Posted by Benny27 | Report as abusive

@behindthecurve – a look at GNP should (I think) account for some of the international exposure companies have. However, I’d be curious if the various tax entities foul up some of the country accounting, since US GNP & GDP are very similar. I believe about 40-45% of S&P revenues come from international segments (although these are revenues, not net income).

A better angle would be to look at S&P 500 earnings as a share of GNP / GDP. This chart looks a lot less scary (with the percentage coming in a little below 2007 levels), and fits better with the context of the discussion, rather than all corporate earnings.

Posted by djiddish99 | Report as abusive

In practice, stock, buybacks take shareholder money and buy shares back from management at cyclical highs. Those were shares that shareholders granted to management at much lower levels. Rinse, lather,repeat.

Give me a growing dividend stream anyday. Also make it more difficult for management to do those all to common ego building value destroying acquisition binges.

As people pointed out up thread, huge component of historical long term equity gains come from the dividend reinvestment in big bear mkts.
Felix and Matt are really behind the times in there thinking on this. I expect more from them.

Posted by gman1 | Report as abusive

Felix, I think you miss Matt’s point – he thinks that companies should return cash to society as a whole, if necessary by sinking money in pie in the sky projects that mosly fail but could have enormous payoffs.

I agree with him.

I have to admit that in general, I would like companies to pay more in dividends rather than sitting on the cash. I prefer dividends to share buybacks because it’s actual money. A good chunk of my retirement portfolio is in dividend paying stocks. But the thing is that when companies pay dividends, the most of the money goes to the 1%. They may buy yachts or whatever and stimulate the economy. But perhaps it would be better for society as a whole if companies did what Matt is suggesting.

As to dividends or buybacks: dividends go to the people who currently own the shares. Buybacks, if the company actually follows through, rewards anyone who owns the shares or sells them shortly thereafter. I am not saying that companies should set out to screw the Icahns of the world, but they should favor dividends because they reward the current shareholders. Of course, shareholder is a pretty nebulous concept.

Posted by weiwentg | Report as abusive

@KenG, I focus on companies with a strong cash flow that use their cash sensibly. I’m happy with share repurchases as part of the picture, at least when the P/E is low. I’m less happy with share repurchases as the exclusive use of cash. You are right that most share repurchases are poorly timed, executed when business is good rather than when the stock is cheap.

No proof, but I suspect that share repurchase programs don’t typically create supply scarcity in the market. Liquidity is far too high and the volume repurchased is too low.

Given a sufficiently low valuation, a strong share repurchase program will force prices higher regardless of speculator psychology. Try totaling up the amount that IBM has spent on share repurchase over the last decade — if it were still selling at $80/share, how many shares would still be outstanding? Per-share earnings would be through the roof, and the speculators would eventually take notice.

Posted by TFF | Report as abusive

Also should note, KenG, that the approach I’ve been using for the last ~5 years stopped finding bargains this year. The dividend growth stock meme may have played out? My portfolio at this point is pretty neutral with respect to the S&P500.

Posted by TFF | Report as abusive

TFF, you’re making the point that only some share re-purchases help shareholders. For every IBM there are ten companies not as successful (okay, that’s a guess).

I don’t mind as much if stock re-purchases are described as attempts to drive share price, but calling that practice “returning cash to shareholders” is pure corporate speak BS. None of that money goes to shareholders.

You stopped finding bargains because after five years, lots of other people wanted the same thing you did. Companies need incentives to become vehicles for making profits by operating their business (i.e., selling things or services), rather than trading assets without adding value. If the tax system penalized hoarding and instead rewarded distribution of profits (dividends), profits would be re-circulated, and stock prices would have more correlation with their performance. And then you could own stocks based on whether you think they will make money, not whether you think other people will want to buy their stocks.

Posted by KenG_CA | Report as abusive

Agreed, KenG. Most repurchase programs are only marginally useful. Some are a waste of money, or an attempt to sanitize excessive options grants to executives.

And yes, I understand why I stopped finding bargains. It was in some sense a bet on slow economic growth, an environment in which “financial engineering” has greater impact than attempts at operational investment (for most companies). As you say, the strategy attracted widespread attention in 2011 and has since pretty much played out. So time for conventional portfolio management unless/until the bargains return.

Posted by TFF | Report as abusive

What helps the most, honestly, is not to focus on the shareholders, but to focus on the people who drive economic activity; their employees. Rather than engaging in buybacks and increasing dividends, they SHOULD be raising wages and letting money flow into the economy rather than hoarding it.

Yes, it’s important to ‘reward’ investors by giving them a return on their loan. But it should never EVER be the primary focus of any company. The primary focus of every company should be to provide goods and or services at a reasonable cost.

Posted by Burns0011 | Report as abusive