"Do you know the only thing that gives me pleasure? It is to see my dividends coming in." - John D. Rockefeller

Encouraging earnings results continue to roll in. While most analysts have pulled down growth expectations from the fast pace of recent years, the consensus outlook continues to anticipate reasonably strong earnings growth this year of 10% or so. As such, not too many corporate executives and analysts are complaining of "poor earnings visibility" these days. If earnings expectations indeed are correct, they'll provide some support for recently-sagging equity prices. But, as most observers know, consensus earnings forecasts are rarely on target. If so, where can one look for more accurate direction? It's an important question as a successful asset mix strategy partly depends on the answer.

One of the most reliable earnings models we have used over the years is based on dividend trends. Why has it proven so reliable in determining profit trends? Most probably because "dividends don't lie." And, lately they have again been speaking.

It's worth a listen. The last time dividends spoke, we were alerted to a strong corporate earnings rebound as of late 2002. It was one of the factors that caused us to remove our deep equity underweights in early 2003 ... in time to participate in the equity market recoveries since that time.

However, dividend patterns have been divining new directions of late. The message? As we will show, corporate earnings growth is headed sideways to down into the end of 2006. Stock markets may already be taking notice. How is it that dividend trends can provide such an "inside tip"?

The Pleasure of Dividends

But first, a quick primer on the verities of dividends. They bring more benefits than just their "coming in," to quote the words of John D. Rockefeller. Following the many corporate scandals and cooked accounts of recent years, the straight hard fact of a cash dividend payment is more than appreciated. As they are mostly paid in cash, what you see is what you get. They're not subject to revision and adjustments - whether seasonal, hedonic, or inflation-related - or aggrandizing definitions that can gloss the profit statement. And because of this verifiable nature, dividend trends have predictive qualities almost like an "inside tip." Managements' decisions with respect to dividends can provide an insight into expected profit trends.

This "tip" works, based upon on a few presumptions: Firstly, that managements loathe cutting dividends. To do so is perceived as a negative event, perhaps triggering a severe decline in the share price, and not incidentally, the value of executive share options. Therefore, companies only tend to raise dividends when they are reasonably sure of a sustainable earnings base. In this way, they seek to avoid the embarrassment of a dividend cut even in a cyclical downturn. It's this vested behavior that can signal "insider" profit expectations.

Whenever dividend increases start to trail earnings improvements, it implies that managements do not regard current earnings levels as sustainable enough to ensure a permanent dividend increase. In environments like this, then, the dividend payout ratio declines. Anytime this has happened over the past half century, and earnings momentum has decisively peaked, earnings have usually fallen to a level compatible to dividend levels within two years. Sometimes, the earnings declines have been much greater.

Dividends Pay Lessons

Some recent examples will illustrate this predictive effect of dividends trends. Back in the late 1990s, dividend policies had been warning of an impending earnings slump already in late 1997. A solid signal indicating an ensuing earnings downturn occurred in mid-1999. Unfortunately, investors and analysts only came around to recognize this reality as late as 2001 ... much too late to avoid the maws of the biggest stock market declines since the 1930s. Then in late 2002, after a period of falling earnings - even as dividend payments remained relatively steady - a reverse signal was triggered. In The Global Spin dated February 17, 2003, we reported the good news of "an earnings upturn of roughly 20% within 2 years from present levels." It proved to be a timely signal and even eerily accurate. Pre-tax earnings actually recovered 19.5% over the following two years from the levels of fourth quarter 2002. All calculations are based upon quarterly figures published in the Federal Reserve's Flow of Funds Report. Although the implied accuracy is accidental, it is worth noting that earnings of the third and fourth quarters of last year showed very little growth over year-ago levels - 1.2% and 2.1%, respectively. Given the continuing strong earnings reports, that may be surprising. That's down sharply from a 17.% rate the year before, and in inflation-adjusted terms is little more than zero.

What are dividends trends predicting now? In our interpretation, an earnings decline of a minimum of 10% from fourth quarter 2004 levels by the end of 2006. If this is correct, it suggests a reason why stock markets are already weak and may yet fall further. At the very least, equity markets will be hard-pressed to match bond and cash returns for a time. That supports our cautious asset mix currently.

However, we did say that we "interpreted" a signal from recent dividend trends. The reason we say that is because even the cash dividend payment is no longer the "honest tattler" it used to be. In the recent decade, dividend policy has increasingly become a tool to float equity markets rather than simply a means to return income to shareholders. As the culture of the "executive share option" began to build as of the early 1980s, and then really took off in the early 1990s along with the EVA ("economic-valueadded") theory and other such similar ideas. The dividend payment gradually took on a new purpose. As such, the time-worn verities embedded in the dividend payment have since become muffled. As such, we have needed to amplify the signals with a hearing aid lately. Let us explain.

The Dividend - New Wealth Generator

Managements seem to have changed their perspective on dividends, whether paid cash, in stock or by way of stock-buybacks. Apparently, dividends are no longer primarily seen as a means to gradual wealth accumulation, but as a device to prop up share prices and achieve quick wealth effects. As mentioned, this shift becomes noticeably apparent as of the early 1990s.

After many decades of level patterns, US dividend payments abruptly started to rise as a percentage of net national income as of the early 1990s. (See Figure #2 below) From 1952 to 1990, dividend payments of the US business sector averaged 2.7% of total national income (NI) with little variance. According to other data sets, a similar level also prevailed in earlier times. But since 1990, payments have experienced a steady surge to a peak of 4.6% of national income in 1998. Dividend payments have remained at that approximate level from that point (apart from the temporary peak of 5.2% of national income in the fourth quarter of 2004 which was primarily attributable to the large one-time Microsoft dividend). All in all, dividends have been boosted 70% above the prior post-war norm. Yet, all the while, the share of corporate after-tax earnings in NI has been trending down over the past four decades, apart from the normal cyclical upturns.

Following a cyclical peak of 7.6% of US NI in the third quarter of 1997 (some seven and a half years ago) after-tax profits of 6.9% of NI are still nowhere near these highs. A general downward trend in the share of corporate earnings has been evident since the late 1980s.

Figure #1

The point we make is this: This recent bias to use dividends to prop up equity prices has distorted the predictive quality of dividend payment trends. As such, we've had to make adjustments in our models to take account of this effect. Having done so, we're still able to hear what dividends are trying to say ... although a little more faintly. And, as "dividends don't lie" (as the popular saying goes) we have reason to take note. Corporate earnings are headed for a downswing over the next four to six quarters. The decline could be as mild as 10% or much more severe if the financial bubble completely collapses. If so, that would release much air out of the financial sector and the infamous "carry trade." As is well known, financial earnings now make up over 40% of corporate earnings and more in the US.

Can You See the Bubble From Here?

Hardly a day passes without a new report being released that concludes that there is no bubble and that there has been no bubble - whether in debt issuance, financial paper, or real estate. And to remember, that only represents the view of those that think that bubbles can actually be discerned in the first place. Many assert that money manias can only be diagnosed after the fact. We hardly agree with either crowd. For one, anyone with a bit of classical economic training will possess the diagnostic tools to identify a financial bubble. And, assuming the latter, the signs of a continuing financial bubble could not be more pronounced. While we won't take the time in this issue to provide comprehensive proofs for these statements, consider the parallel trends evident in the corporate sector that align with a massive credit inflation that has boosted the share of consumption (in GDP), overall indebtedness, and crucially, a gradual collapse in capital investment.

Figure #2 on this page clearly shows a relationship between rising dividend payments and lower corporate retained earnings as a percent of national income. Actually, it only follows that this would be the case. But, if nothing else, it clearly illustrates the obvious that most analysts miss. High dividend payouts - whether in the form of stock dividends, cash or stock buy-backs - are theoretically a sign of trouble, not imminent prosperity. Companies that pay back capital or pay high dividends historically would do so for the reason that no better investment opportunities were available. Lower retained earnings overtime mean both less opportunity and reduced means for investment. And, those that remember the underlying theory behind the workings of capitalism, will agree that high investment - productive, income-generating investment - is the road to profits and prosperity. In theory, high dividend payouts and capital distributions for the business sector overall is a sign of future trouble for national prosperity.

Of course, a company executive laden with thousands of share options or a heavily invested portfolio manager might be persuaded to see it differently. From a stock market perspective, there couldn't be anything more bullish than rising dividend payments and stock buybacks as a per cent of national income during a time of falling interest rates. Figure #3 shows how conveniently these two trends coincided during the 1990s. It helped generate the stock market boom of the 1990s, one of the biggest in history. Of course, that too is history. But, what of the future? Imagine that the trends depicted in Figure #3 were the exact opposite - a period of slumping dividends as a percent of GDP and flat or rising interest rates. Could that happen?

Figure #2

Figure #3

Dividends tell us the truth on that question. We only have to listen to them. Dividends continue to remain high as a share of national income and that means that future profit trends will eventually prove to be dismal. That must be true for either of two reasons: 1. Profitable capital investment opportunities by implication must be few and far in between, and/or; 2. There isn't sufficient income retained to drive healthy levels of capital investment that will drive future profit upturns. Either conditionds undermine future profit potential for the business sector as well as the nation overall.

And, not to forget, dividend patterns have recently spoken on another matter as we have already concluded. A cyclical earnings downturn is already in progress.

Three Important Conclusions

What do dividends imply for the stock market outlook? To be sure, dividend yields themselves are hardly much more attractive than they were. Yields still remain far below the averages of the past 50, 20 and even 10 years.

But as outlined, what's more noteworthy is that dividend payments remain unsustainably high relative to national income. That has negative implications for future profit potential.

Recent dividend trends also counsel one additional insight. A corporate earnings downturn is likely before the end of 2006. That reality is still not fully reflected in current stock market levels. While it may take some time for markets to recognize this - after all, current earnings reports are still seen as robust - it argues for a cautious stance on equity market allocation in our managed portfolios.

Don't blame us. We didn't say it. "Dividends don't lie."