Do dividends matter?
Dividends matter, says James Montier of Dresdner Kleinwort Benson. A lot of investors tended to loose sight of this during the 1990s bull market - and it is easy to see why. In the period 1995-2000, over 80% of the return achieved in the US stockmarket was generated as a result of the higher multiples investors put on company earnings. But this was totally unprecedented. If you look at returns from the US market over the long run since 1900, you will find that 66% of the total return was generated by the dividend yield. And if one looks at the period 1950-2000, one finds that 72% of the returns were delivered via dividend yield. The US is not unusual in this, says Montier: “Across nine major global markets, we find that an average 65% of total return has been generated by the dividend yield since 1950.”
Yet in recent years, dividends have been out of favour with academics and investors, noted The Economist. Investors have argued that the yield on a share is irrelevant. A low yield may simply show that firms are retaining earnings for reinvestment or to return cash to shareholders in more efficient ways, such as share buybacks. These arguments were backed up by a certain amount of academic research. Nobel laureates Franco Modigliani and Robert Miller taught that dividend policy is indeed irrelevant. “It makes no difference to investors whether a company pays out all its profits to shareholders or keeps and reinvests the lot. (In practice, however, tax systems skew companies’ decisions. America’s tax system has led companies until this year to favour capital gains over dividends).”
Certainly, a refusal to pay dividends seems to have done no harm to the performance of one of America’s most successful companies, said Tony Glover in The Business. An investment of 1,000 shares in Berkshire Hathaway, Warren Buffet’s investment vehicle, would have cost $20,000 in 1965 and is now worth $75m. But Berkshire Hathaway hasn’t paid a dividend since 1967. Berkshire shareholders, says Buffett, “don’t care” about them - although that may change in the light of the new tax breaks introduced for individual investors.
(Article continues below)Advertisement
But this modern obsession with earnings at the expense of dividends is a mistake, says Philip Coggan in the Financial Times. The historical data suggests dividends are more important than investors think and is a good indicator of future earnings growth. According to a recent study of the Standard & Poors 500 index over the period 1950 to 2000, future earnings growth is positively correlated with the payout ratio: when the payout ratio was high, so was earnings growth. The payout ratio in the US edged up to 0.47% by June, just below the long-term average, from a low of 0.3% in 2000, but is now deteriorating again (as earnings rise faster than dividends in the recovery). This suggests that investors should expect below average real earnings growth over the next ten years.
The signs are that dividends are coming back into fashion, says Clare Gascoigne in the Financial Times. In the UK, dividend income grew in 2002 for the first time since 1998, and most analysts are expecting a further rise this year. But buying high-income shares takes “a bit of serious homework”. A high yield could indicate a problem with the company, rather than “an overlooked gem”. Check the cash flows and balance sheets for financial strength and look at dividend cover (which shows how much money a firm has to cover its dividend payout) to ensure you are not “picking a dog”. If an investment is paying above-average yields, it is worth asking why - and what additional risk is being taken on in return for the high yield. If the yield on a share is higher than the yield on long-term gilts (currently 4.95%), ask yourself whether it is too risky.
