WASHINGTON, Jan. 8 (UPI) -- (In view of the president's stimulus proposal Tuesday, UPI is refiling a Bear's Lair column originally published Nov. 27, 2000, which discussed the desirability of focusing attention on dividends rather than capital gains.)
As the tech bubble deflates, analysts are returning their attention to "value" stocks, in which the companies actually have earnings, and price-earnings ratios, by the standards of the tech bubble, appear more reasonable.
Of course, the earnings investors are being asked to buy may not be real, and dividend yields remain tiny. But then Modern Portfolio Theory says dividends don't matter, so investors can just load up on the "value" stocks and sleep quietly at night. Can't they?
No. What are currently regarded as "value" stocks are trapped in a bubble almost as over-inflated as the tech stock bubble, compared with traditional standards of valuation.
Modigliani and Miller, in their Nobel Prize-winning articles of 1958 and 1961, suggested that dividend levels were irrelevant to investors. Investors who wanted income could simply buy a mixed portfolio of bonds and stocks, or could sell a portion of their stock holdings each year, reflecting the underlying earnings on the companies concerned.
For retail investors, this theory was always less than realistic. Retail investors use their portfolios to build net worth for their sunset years, then live off the income from their portfolio. They need a steady, predictable stream of income, preferably one which will increase with inflation or a little faster, and which will not fluctuate violently with the stock market. They also need to be sure that their income will last them out, and that they are not eating into capital. Selling stocks to create cash flow suffers from the problem that more stock is sold in market downturns; it is the opposite of dollar-cost-investment averaging.
It also suffers from the problem that investors don't know how much of their portfolio to liquidate, because they don't know how long they'll live. Guess wrong, and they either have an extremely penurious old age, or they live more frugally than they need to and pass an excessively large estate to possibly ungrateful heirs and a definitely ungrateful government in the form of "death taxes." For this reason, even those investors with faith in company management are attracted by dividends, although clearly endlessly repeated 20 to 25 percent per annum capital gains would be even more attractive.
While much of their work has proven to be inapplicable in real-world financial markets, Modigliani and Miller's denigration of dividends has endured. There is one good reason for this: dividends are subject to double taxation, since the company paying the dividend has already suffered tax on the underlying earnings.
There is also one very bad reason. The value to company management of a corporate stock option schemes is based on the unadjusted market price of the company's shares, measured against a fixed option exercise price. Hence a company paying a large dividend is reducing the value of management's stock options. It is not surprising, therefore, that the dividend payout rate (dividends/earnings) on the S&P 500 has gone from 54 percent in 1990 to 29 percent in 2000.
Management is retaining more earnings, buying back stock, and leveraging the company, all of which activities increase reported earnings per share and the stock price, thus raising the value of its own stock options. It is not clear that any of these activities are in the interests of the ordinary stockholder.
Retaining earnings reduces stockholder income, diverting resources from the stockholder to management. Only if a stockholder has blind faith in management should he think this desirable.
Buying back stock reduces the stock's liquidity. More important, since stock buybacks increase when times are good, the stockholder will find the company buying back its own stock when the price is highest, reducing investor returns below what they could achieve directly.
Increasing leverage increases returns while times are good, but it also increases the risk of corporate failure, thus imposing an agency problem. In the event of a bankruptcy or corporate restructuring, stockholders will lose much or all of their investment while management, even if they lose their jobs, will be well protected by "golden parachutes."
For all these reasons, investors should prefer cash in the pocket to retained earnings.
Management stock options give rise to a further problem in that reported earnings themselves may be bogus. This column has noted that Cisco's earnings disappeared in every year from 1996 to 1999 when the value transfer to management from stock option exercises (which, owing to the eccentricity of current Generally Accepted Accounting Principles, doesn't appear in the Income Statement) was taken into account. Similarly, Microsoft's 1999-2000 earnings disappeared (the value of the stock option wealth transfer to management was $13.9 billion, compared with net income of $9.4 billion), as did Microsoft's 1998-1999 earnings (wealth transfer $8.5 billion to management compared with net income of $7.8 billion), and its 1997-1998 earnings (wealth transfer $4.7 billion compared with net income of $4.5 billion.)
Of course, these excessive stock-option value transfers are the result of the great 1995-1999 bull market, in which stock prices never dropped. In a normal market, or a falling market, the payout to management from stock option exercises would be much lower, or even negative. However, who imagines in such a case that key management would stay with the company if they were paid only their base salaries? In such a case, management would of course demand re-pricing of their stock options, so that gains were achieved at lower stock prices, as well as substantial increases in their base salaries. Unless the economy or the company's prospects were really dire, management would no doubt get both. It would be at this moment, with the extra charges flowing through the Income Statement, that Generally Accepted Accounting Principles and reality would once more converge, giving Wall Street analysts a very nasty surprise indeed.
General Electric, physically although not in terms of market capitalization a much larger company than either Cisco or Microsoft, was less extreme: In 1999 the stock option wealth transfer to management was $1.95 billion, only 18 percent of net income of $10.7 billion. Interestingly, General Electric also paid a genuine dividend, at $1.46 per share 45 percent of diluted earnings per share. If, therefore, of the three largest-capitalization companies in the world, a reasonably conservative long term investor had to choose one, General Electric, with less than 20 percent of its earnings going to management in the form of stock options, and a dividend of almost half earnings, would be far and away the soundest choice.
For retail investors, Modigliani and Miller's "dividends are irrelevant" theories are like Bauhaus architecture, a bizarre mid-20th century aberration over which, thankfully, time is now drawing a veil.
Of course, investors who look at company dividends, rather than relying on the greater-fool-theory valuations brought about by ever-rising stock prices, are able to compare dividends with returns on other securities, and price stocks accordingly. In this context, it should be noted that the theory whereby stock dividend yields could remain perpetually below bond yields, because of inflation, received a huge blow a few years ago when the U.S. Treasury brought out inflation linked bonds. Investors now have an inflation-linked benchmark yield to which they can compare their dividend paying stocks. For the S&P 500 Index, this comparison might run as follows:
The dividend on the S&P 500 Index, if the companies therein returned to their normal historic payout levels of around 55 percent, would be about $30. A rational investor would believe this dividend to be effectively inflation-linked, but would demand a modest yield premium over index-linked Treasuries, because there are two additional risks: the risk of corporate over government obligations (which is currently about 1.5 percent, but normally around 1.0 percent), and the risk of stock over debt returns, which is compensated for by the possibility of real growth in stock returns. Hence, as index-linked Treasury bonds today yield 3.83 percent, it would be reasonable for an investor to demand a dividend yield of 4.83 percent on the S&P 500. With the "normalized" dividend of $30, this translates to a level for the S&P 500 of 621 -- compared with the current level of 1358 as I write. You can juggle these figures a bit, but the underlying point remains: for a prudent bear, wishing to achieve a steady and if possible growing real dividend income, the S&P 500 Index is currently at least twice its proper level.
There is also a tax reform issue here, the double taxation of dividends in the U.S. Most other countries, although having higher corporate and personal income tax rates, do not do this. The United Kingdom, for example, has a system of Advance Corporation Tax, whereby dividends paid can be deducted from next year's taxable corporate income, and are deemed to have paid tax at a standard rate.
The next Congress could thus return power to stockholders from management by a simple reform: it could allow dividends paid in relation to a year's income to be deducted from corporate income before tax is calculated. In this way, dividends would be taxed once, at the recipient level, and retained earnings would also be taxed once, at the company level.
With the lower taxation of dividends, their gross level would rise, by a factor of approximately 100/65 (reflecting the current 35 percent corporate tax rate), which, on the yield assumptions above, would raise the appropriate level of the S&P 500 Index from 621 to 955, only 30 percent below its present level. Phew!
--
(The Bear's Lair is a weekly column which is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that, in the last 10 years, the proportion of 'Sell' recommendations put out by Wall Street houses has declined from 9 percent of all research reports to 1 percent. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)
