PsyFi Search


Saturday, 9 January 2010

The Psychology of Dividends

Dividends Can’t Matter - Rationally

One of the more amusing failures of classical economics is its inability to explain why companies pay dividends. The Miller and Modigliani (1958, 1961) synthesis shows that rationally dividends are irrelevant to a corporation’s valuation: after all, if you give investors some of the company’s earnings then the company should be less wealthy to the same degree that investors are more, so there’s no net gain for shareholders.

Which is all nice and theoretically sound in the hermetically sealed world of rational economics but, unfortunately, when companies cut dividends their valuation usually drops, often quite dramatically, and when they raise them significantly the opposite happens. As usual, out in the real-world, it’s what people do, not what economics dictates, which rules.

Give and Take

The rational analysis of the irrelevance of dividends is perfectly grounded in straightforward economics. If I own a share which gives me access to $2 of a company’s earnings and the company gives me $1 directly then the company’s share price should drop by $1 and I will be $1 personally better off (minus taxes). So all things being equal a dividend should be irrelevant.

In fact dividends may well be damaging to a company’s prospects. If instead of giving me $1 the company invested it in product innovation, or marketing or an earnings enhancing acquisition (no laughing, please) it will be able to grow its earnings more quickly. I, the shareholder, benefit with a more rapidly increasing share price. If I want some of this for my own personal consumption I sell some stock.

Of course, large and mature companies may not be able to find sufficient earnings enhancing opportunities and so, all things still being equal, may decide that returning excess earnings to shareholders is the best thing. Again dividends aren’t the only way of doing this, although as we discussed in Buyback Brouhaha the main alternative – stock buybacks – is shrouded in managerial deceit and accounting opaqueness in a way that dividends aren’t. Although buybacks may be more tax efficient, depending on your taxation jurisdiction.

Dividend Signalling

So there are valid reasons for dividends, even if the classical view states that this won’t benefit shareholders directly. At least returning cash to shareholders allows us to redeploy these into other opportunities with better earnings prospects – although, as has often been pointed out, we can do that by selling one company’s shares and buying another’s.

Overall, then, it’s not at all obvious why companies should be overly concerned about dividends. However, they are, because they spend a great deal of effort manipulating dividends and giving indications about future dividend policy. And they’re right to do this because surprises in terms of dividends tend to cause significant share price movements in spite of what classical economics tells us should be the case. One of the possible explanations about why dividends persist in spite of economists saying they shouldn’t is their signalling effect.

So, a change in dividend policy may often indicate a change in the company’s fortunes. A cut in dividends will often signal reduced earnings – although it sometimes indicates that there are better earnings enhancing opportunities around. Similarly an unexpectedly raised dividend will often see a share price surge – even though this often indicates that the management have run out of ideas about how to deploy their spare cash, which isn’t exactly a positive sign.

Desperate Dividends

However, to argue that the only reason dividends exist is to give managements a way of showing their confidence or lack of seems, well, desperate. After all management could rather more simply tell us directly that they have low visibility of future earnings – they’ve been doing that rather a lot recently. No, the real explanation seems to be that investors often prefer dividends: so why might investors have a preference for higher taxed dividends over internally reinvested earnings?

Well, firstly, returning free cash to shareholders removes from management the temptation to waste it on pet projects. It also has the rather odd effect of disciplining managements because they will more often have to raise additional capital in the market. Strange though it is, companies that pay out dividends are often simultaneously tapping the markets for additional capital. That additional capital costs the company money in fees, dividends cost the shareholders money in taxes and the total result is a significant reduction in earnings available to the company’s owners.

We live in a strange world.

The Psychology of Dividends

There are also multiple behavioural reasons why investors might prefer dividend paying stocks over non-dividends. Firstly, receiving an income stream means that investors don’t need to sell stock to receive an income, which can often be a source of regret (which we discussed in ... err ... Regret) if the company subsequently does well. Of course, investors could have reinvested their dividends in the stock but this is a sin of omission, as opposed to a sin of commission, and is far more easily ignored, as suggested by Shefrin and Statman.

Secondly, the problem of self-control that we discussed in Retirees, Procrastinate at Your Peril is far easier to manage if investors decide to spend only their dividends. Although the research suggests quite strongly that the only stock market growth available for long periods is through dividend reinvestment investors will often spend the “interest” on their dividends anyway. By avoiding any sale of capital it’s easier to control the urge to spend the lot. This is Mental Accounting again, of course.

Finally there’s a case that a stock paying a high dividend today is perceived as a better bet than one that may provide greater earnings and price increases in future. This is known as the “bird-in-the-hand” fallacy.

Residual Dividend Policies

One of the odder findings in research on dividend paying stocks is that those companies which follow a so-called residual dividend policy – essentially paying out their entire free cashflow to investors as dividends – are generally more financially sound than companies paying out less of their earnings. Explaining this isn’t easy, but it’s possibly worth noting that these higher dividend payers tend to be larger and find it easier to raise external capital.

It’s certainly a counter-intuitive idea for investors to look for larger companies with lower free cashflow, but those companies with residual dividend policies tend to have longer term outlooks than others. In essence these companies don’t actually aim to run with low free cashflow but instead set dividend policy based on expected long-term earnings, rather than adjusting based on the short-term winds of fortune. So perhaps this finding is simply stating that those companies whose managements take a long term view and want to maintain a stable shareholder base are likely to be better aligned with their long-term owners than others.

Note, though, that "alignment" means understanding shareholder psychology and playing to it, attending to the lessons of behavioural finance. This isn't necessarily the same as maximising shareholder value as promoted by more orthodox economic theories. We'll revisit this issue, soon.

Dividends Are Not Enough

However, it’s fairly clear that deciding on an investment policy purely on the basis of dividends without regard to the nature of the underlying corporation is a pretty stupid idea and one that’s founded in behavioural fallacies. The behavioural tricks and twitches that make people adopt this kind of approach regardless of the underlying robustness of the institutions involved is simply another facet of the psychological blindness that many investors have with regards to stockmarket investment.

In the end, there’s none so foolish as those that are blinded by their own behavioural failings. Most of us can recognise the psychological problems of stockmarket investing in others yet the majority of people will still fail to acknowledge their own issues. Dividends are simply the tip of a very a large problem. Still, on the positive side, well managed dividend payers are amongst the best bets in the market. Just don’t forget to reinvest the dividends while you can, otherwise you’re spending the majority of your future wealth.

Related Articles: Real Fortune Telling, Buyback Brouhaha, Debt Matters, Don't Overpay for Growth


  1. My thought is that investors should prefer to receive a portion of their payment for investing in a company through dividends because getting paid now is always better than getting paid later. And smart companies should realize that they are better off attracting the type of investors who like dividends because the companies benefit from having smart investors find their operations. So I take it as a definite plus to see that a company pays a good dividend.

    The other side of the story is that I take it as a definite negative to see that a company does not pay a dividend. So long as there are dividend payers to invest in, why not direct your money toward those?


  2. Tim,
    Excellent blog.

    Of course in a rational world payment of dividends is irrelevant. But, as Dan Ariely points out, we are predictably irrational. Rob Bennett is quite correct to say that a bird in the hand is worth two in the bush, just in case.

    Although it may not be fair, or even accurate, it is a reasonable working assumption for an investor (the principal) to assume the directors of the companies he is in invested in (the agents) are greedy liars and the sooner he can can get some money back the better.

    Companies exist for two reasons. Create cash for their owners and egos for their managers. Paying dividends boosts the first and reduces the opportunity for the agent to do the latter. Berkshire Hathaway, which has never paid a dividend, might be a good example.

    It is arguable that investing is simply about three things:
    compound interest
    maximising dividends
    minimising fees

    That is the philosophy we follow at The Munro Fund, which simply tracks the forecast gross cash dividends of companies in the FTSE 350.

  3. Interesting, except that it occurs to me that in a world free of dividends, stock ownership would never result in a cash receipt. Then, it would not make sense for anyone to buy the stock, thus it would never appreciate, removing the opportunity to cash in via capital gains. A claim to a portion of a company's earnings is merely theoretical if cash is never received as a result of that ownership.

  4. Suppose I told you I was opening a business, for instance a restaurant, and that I was looking for capital. For the right price, you can own half the business.

    You would probably inquire how much income you could expect from this investment.

    Well, I would explain that you that according to our contract you will get no income (it will be reinvested into the business). However, you can sell your stake in the restaurant at any time to someone else, who also will never receive income.

    Oh, and if you like the sound of that, I've got some real estate on the moon that might also be of interest to you.

  5. Anonymous and Parker Bohn have beaten me to it. You don't need to be irrational to have doubts about non-dividend paying companies, you only need to appreciate that others may be "irrational" - and that makes other people's supposed irrationality entirely rational. Dividends are what prevent shares becoming an unconvertible currency of their own, or a currency which indefinitely floats free of any hard mechanism to force a rerating.

    Look at investment trusts and smaller companies that are undervalued on any sensible measure because they don't pay dividends. I've invested in some of these because they provide substantial opportunities. But I wouldn't invest in a company that had a stated policy of never paying a dividend (even after it became profitable in the future), any more than I would buy into a Ponzi scheme.