Deliberating on dividend policy.
FOR THE VAST MAJORITY of larger public companies, dividend policy is an important component of maximizing shareholder value. While dividends have always been an important part of shareholder returns, dividend yields have historically been much higher than they are today. Since peaking in 1979-80 at a yield of 5.7%, dividend yields have gradually drifted lower: in 1990, the dividend yield of the S&P 500 was 3.4%; in 1995, 2.9%; and in 1999, 1.4%. Furthermore, with the rapid rise in stock prices, in recent years, dividends have comprised a much smaller portion of total shareholder returns than they have historically. For example, over the past five years, yield has represented 8% of the return on the S&P 500 versus 34% for the 1926 to 1999 period. While current high stock market prices have somewhat reduced its importance, dividend yield is likely to grow again in prominence when share price appreciation returns to more normal levels, or if stock prices retreat.
While it's difficult to generalize about dividend policy because it is usually very company-specific or industry-specific, some general observations are possible. While current earnings and expected earnings growth drive a company's stock price, effective dividend policy can influence its trading multiple within a range. Dividend policy's most important uses are to: (1) return excess cash to shareholders; (2) effectively manage the company's cash needs and capital structure; and (3) credibly signal shareholders about future earnings prospects. The recommendations herein should be viewed as a starting point.
It is our view that dividends are the primary vehicle for returning cash to shareholders. A company should choose a payout ratio that directs the majority of its free cash flow to shareholders as a dividend. In doing this, a company should avoid a dividend payout that is too high, in part because a high payout ratio reduces its financial flexibility and increases the risk of a ratings downgrade and/or dividend reduction. Cuts in the dividend also should be avoided. Dividend increases should signal confidence in the future; this is especially important in times of earnings weakness, and less important during periods of strong earnings. If a company can afford a dividend, in most cases a nominal one should be paid, expanding the stock's appeal to the broadest possible group of shareholders. Irregular special dividends and stock dividends should be avoided.
Dividend policy should be viewed as part of an integrated financial plan. Other than for firms that trade primarily on yield, a company's stock price largely reflects its current earnings and future earnings growth rate, not its dividend policy. Shareholder value is maximized through an effective investment strategy, financed by an optimal capital structure. As shown in the accompanying exhibit, we view both dividend payments and share repurchases as a byproduct of these strategies and therefore an important, yet residual, decision.
Use the dividend as the primary vehicle to distribute cash to shareholders. Once a company's operating plan has been appropriately financed, dividends should be the primary source of cash flow paid out to shareholders. As a payment that shareholders can largely depend on receiving, dividends add stability and bond-like qualities to a company s stock price. This is generally positive, as it somewhat lessens stock price volatility, and can slightly lower a firm's cost of equity.
Another important consideration is that companies in the same industry often will generate similar levels of excess cash flow, causing them to grayitate to similar dividend payouts. As portfolio managers select among this group of potential investments, it is important that a company is not viewed as paying out a substandard dividend or viewed as "holding back" from committing to regular cash payments to shareholders.
Use the dividend to appropriately signal shareholders about future prospects. Dividends play an important role in signaling shareholders about management's confidence in future earnings prospects. Firms that base dividend decisions solely on past earnings (e.g., pay out a fixed percentage of income) lose much of this signaling benefit. Ideally firms could pay out a portion of anticipated earnings, although the uncertainty of future earnings makes this a risky strategy. We recommend that most firms combine the two in determining their annual dividend.
To the extent that a company's earnings have "permanently" reached a new level, dividends should be increased. However, since the market will focus primarily on earnings growth in years of strong earnings, firms can temper their dividend increases in these years (lower payout). In average years, a company's earnings and dividend increases should be similar (unless the firm is managing its payout up or down long term). In years of weak earnings, assuming the company is positive about the future, the dividend can signal confidence. Actions can vary from holding the dividend steady in the face of an earnings decline, to increasing it at a slightly faster rate than earnings growth. The goal should be to committing long term to maintaining a comfortable total dividend payout ratio.
Avoid a high dividend payout ratio that provides little flexibility. Since firms wish to avoid a dividend cut or a downgrade in their credit rating, and maintain some flexibility over free cash flow and dividend changes, it's typical to set the dividend payout ratio below the likely full payout ratio; how far below depends on the underlying stability of earnings and cash flow, and the stock price value attributed to yield (versus share appreciation). While this "comfort" level varies for most firms, it would probably be between 60% and 80% of the maximum dividend payout ratio. For cyclical firms, this percentage would be the average level throughout the cycle. (Obviously firms that trade largely on yield should distribute most of their free cash flow to shareholders through the dividend.)
Avoid dividend reductions. Companies should avoid dividend eliminations and reductions for many reasons. One obvious reason is that these actions typically lead to a significant drop in a firm's stock price. With dividend policy often used by companies to signal the market about future prospects, a dividend cut sends a clear negative message to the market. Furthermore, in a pattern that is self-reinforcing, cuts are avoided simply because they are so rare.
To significantly reduce its dividend payment, a company typically must be under severe current financial pressure and have strong concerns about future cash flows. Dividends at most companies, therefore, are viewed as virtually a guaranteed payment to stockholders -- a payment that, as stated, most firms will struggle to maintain. When this payment is cut, it becomes a clear sign to shareholders that management has failed to produce expected earnings and manage cash flows. This fact pattern makes it especially difficult for firms in industries that suddenly require much higher levels of investment or whose business mix is changing from low to high investment needs.
To analyze the market's reaction to a dividend cut, we evaluated a sample of 83 U.S. firms that either reduced or eliminated their dividend since 1990. (These were companies with a market capitalization over $500 million that reduced their dividend by at least 25% and had a dividend of at least $0.05 per share prior to the cut.) As expected, dividend eliminations yielded a significantly larger drop in stock prices than did partial dividend reductions: For the 27 firms that totally eliminated a dividend payment, the average stock price decline (net of the market) was about 25%; a partial dividend reduction resulted in an average decrease of about 6%. It should be noted that, while the average reduction in the dividend payment at the "reducing" firms was 54%, the average yield of this group just prior to the cut was a modest 1.8%.
It's also important to note that for both reductions and eliminations, market speculation typically was high that a change in the dividend was forthcoming. Such speculation can continue for a long time, but it typically intensifies prior to the actual dividend announcement.
Finally, this negative reaction to a dividend cut also is evident in analysis of the market reaction to diversified companies that announce a major spinoff or division. At those firms cutting the dividend as a part of the realignment, the average stock price declined over 10% (net of the market) in the firm 90 days following the announcement.
Don't pay special dividends or stock dividends. Firms historically have used special dividends most often to reward shareholders following an especially good year. When a company has a history of special dividends, the market often will speculate at length in anticipation of this payment, companies will struggle to set the appropriate level, and ultimately derive little permanent benefit associated with the payment once it's paid. With share repurchases increasingly common, we fail to see the benefit of a small special dividend. Distributing cash in the form of a dividend increase that is carefully managed, and balancing changing cash flows and capital structure objectives with share repurchases, is a pattern that best builds sustainable value.
We do not attribute any value to dividend payouts in the form of stock. While a few small retail investors may view a stock dividend as having value, the sophisticated institutional shareholders that dominate today's market will not. A firm wishing to increase its shares outstanding, or keep its stock price in a traditional range, should do so through an occasional, appropriately timed stock split, not through a series of stock dividends. Furthermore, the timing of the split can be used to deliver a 3-4% rise in the stock price, a change not associated with small stock dividends.
Appeal to the broadest possible group of shareholders. Many institutional shareholders are unable to purchase stock unless a dividend is paid -- for example, this is required by most pension funds. It's estimated that these investors make up about 20% of institutional investors, or about 10% of total shareholders. To appeal to the broadest possible group of shareholders, companies should consider paying even a nominal dividend. While this is not a necessity for companies with smaller float, it becomes more important as a firm's capitalization grows and it moves into the "large cap" category.
Furthermore, it is not necessarily the case that high-growth firms don't pay a dividend. While firms in need of capital may not wish to pay out any cash to shareholders, for a large high-growth company, a token dividend is usually not a significant payment. Examples of large high-growth firms that pay a dividend include Intel, Motorola, Home Depot, and Compaq. It should be noted that others, such as Microsoft, Cisco, and Dell, don't pay a dividend.
Rick Escherich is a managing director in J.P. Morgan & Co.'s Mergers & Acquisitions Department and head of its M&A Research Group, which is responsible for evaluating numerous issues related to shareholder value and sets policies and procedures covering the valuation of companies and the assessment of various transaction types, analytical techniques, and securities. He joined J.P. Morgan in 1978.
Dividend versus share repurchase decision
Dividends should be the primary source of cash flow to shareholders. While share repurchases are an important part of returning cash to shareholders and maximizing shareholder value, they should be viewed as the secondary vehicle for paying cash to shareholders. (Tax law also prevents companies from just buying in shares and not paying a dividend.)
Value enhancement through cash distribution to shareholders is likely to be maximized by combining: (1) a dividend policy that signals shareholders about long-term prospects; (2) a dividend payout that is capped at a level below its maximum sustainable level (excluding those firms that trade largely on a yield basis); and (3) share repurchase programs that balance cash needs and capital structure objectives. This combination distributes all excess cash to shareholders, and also increases operating flexibility, enhances EPS growth, provides higher returns to taxpaying shareholders, and enhances the value of stock options as a compensation tool.
Increase financial flexibility with share repurchases. Companies gain financial flexibility by reducing their dividend payout and using excess cash flow to buy in stock over time. This is especially important if a company anticipates volatile investment needs or a large cash acquisition. Companies that direct a portion of their free cash flow to share repurchases will find it easier to eliminate these programs temporarily and to maintain a steady dividend policy in the face of large investment needs. Firms already committed to a very high payout ratio may be forced to make unwanted changes to dividend policy in the face of volatile cash flows or the need to finance a large acquisition.
Even for firms that have reasonably stable free cash flows and don't anticipate a large cash acquisition, committing to a dividend payout above 50% or to levels well above those typical for their industry may present problems, should unanticipated circumstances arise. Dramatic changes could occur due to unexpected developments in technology, a "once-in-a-lifetime" acquisition opportunity, or event risk. Event risk could range from an unexpected large cash need due to litigation or environmental problems to dramatic structural shifts in costs or pricing.
While firms can often ride out these periods without a cut in the dividend, a commitment to a larger payout under these circumstances creates dividend decisions that are problematic. The impact of a dividend cut, or a zero-growth dividend, at a firm considered to be reasonably stable generally has negative shareholder value implications.
The required degree of flexibility is an important consideration in estimating the appropriate size of a share repurchase program. In general, this will be the difference between the dividend payment a company is confident in committing to shareholders (and growing over time) versus the maximum level that might be sustainable should there be no unforeseen demands on cash flow. From a practical standpoint, it is important to note that this difference will change from year to year as cash flows and business prospects fluctuate.
Use repurchases to add to shareholders' value, EPS growth, after-tax returns to investors, and the value of incentive options, and to ease periods of selling pressure. As just discussed, the major benefit of supplementing dividends with share repurchases is the flexibility that this provides. But there are other benefits that, in their totality, are not insignificant and play a role in the dividend versus share repurchase decision.
These are summarized above.
While sizable self-tenders can be used effectively to quickly boost a company's stock price (typically for defensive purposes), buy in undervalued shares or illiquid shares, and/or significantly change a firm's capital structure, we view open market programs as usually best-suited to complement dividend payments. Overtime, it would not be unusual for a mature firm to pay out 20% to 40% of its distributable free cash flows to shareholders in the form of a repurchase.
Frederic A. Escherich
Recent trends in dividend policy
With thousands of very different companies independently formulating their respective dividend policies based on their unique investment opportunities, mix of shareholders, and earnings prospects, changes in the area of dividend policy are difficult to identify and tend to evolve very gradually over time. Some of the slow changes that appear to be taking place include:
A trend toward slightly lower dividend payouts. With the emergence of share repurchases and desire by firms to increase their financial flexibility, dividend payout ratios, on average, are decreasing. This market trend is somewhat difficult to prove conclusively, because any analysis is complicated by the influence of cyclical earnings, special charges or gains, the ever-changing composition of the major market indices, and the higher level of goodwill (lower book earnings) resulting from M&A activity. Nonetheless, the annual adjusted dividend payout (dividends as a percent of adjusted net income, excluding extraordinary items/write-offs) for the S&P 500 shows a trend of gradual decline: from 51.4% in 1990 to 35.5% in 1995 and to 32.7% in 1999.
Our sense that dividend payout ratios are slowly decreasing is further supported by numerous discussions on the topic with many large U.S. companies. On average, firms are either satisfied with their current dividend payout, or plan to shave it somewhat overtime. At the extremes, some companies, because of changing industry dynamics and/or investment opportunities, would like to cut their payout dramatically. There appears little, if any, general sentiment on the part of mature companies to increase their payout. As consequences of all these factors, dividend payouts have declined slightly, a trend that we expect is likely to continue.
An increased emphasis on share repurchases. While the dividend payout is declining somewhat, total cash payouts to shareholders -- which includes share repurchases -- appears to be rising. An analysis we have completed of a group of 25 noncyclical Fortune magazine admired" U.S. companies provides clear evidence of the sizable amount of free cash flow paid to shareholders in repurchases: an annual dividend payout for the group of about 40% which has held fairly steady from 1979-1999, versus a total distribution payout that has increased from approximately 47% in 1979 to a rate in the last several years approaching 80%. It should be further noted that the pattern of an increasing total payout would be even more apparent by excluding the takeover-related, very large share repurchases of the 1980s. For example, in 1985 only eight of the 25 companies bought in more than $100 million in stock, versus 19 of 25 in 1998.
The signaling power of the dividend is somewhat less important. Historically, dividend announcements send an important signal to shareholders about management's confidence in future performance. But our sense is that, in today's market, shareholders appear to weight earnings announcements more heavily than in the past and dividend announcements somewhat less. This makes sense in a high-P/E market where yield is playing a less important role.
Therefore, when a company's earnings are at or above the market's expectations, further signaling management confidence with a hefty dividend increase has less impact. As a consequence, companies intent on gradually lowering their dividend payout ratio to increase flexibility can do so without hurting the stock price. However, for dividend changes that coincide with weakening earnings, dividend reductions continue to be seen as confirming investors' fears about current and future prospects and, as in the past, send a strong negative signal.
Frederic A. Escherich
Market reaction to a reduction in dividend payout
Firms trying to reduce their payout would do well to consider the potential effects. The accompanying schematic summarizes our observations on the stock market's response to various reduction techniques employed by companies today.
Obviously the most palatable approach is to grow dividends per share long term more slowly than earnings per share. Coca-Cola is a well-known example of a firm following this strategy; along with other steps, Coca-Cola lowered its dividend payout ratio from 65% to 33% over the period 1983-1997. During this period, Coca-Cola's dividend per share grew at an average annual rate of 12%, versus average annual EPS growth for the same period of 18%.
A stable dividend appears to be acceptable when a company needs to devote all of its increase in excess cash flow to a pressing need -- such as debt repayment, or support in a downturn of an earnings cycle. When a dividend must be reduced immediately, one reduction is always better than a series (similar to write-offs). In the case of stocks that are "yield plays," yield-oriented investors will better tolerate a cut that offers the option of trading their common shares for mandatorily convertible preferred stock.
Frederic A. Escherich
|Printer friendly Cite/link Email Feedback|
|Author:||ESCHERICH, FREDERIC A.|
|Publication:||Directors & Boards|
|Article Type:||Statistical Data Included|
|Date:||Sep 22, 2000|
|Previous Article:||As directors approach the 'zone of insolvency'.|
|Next Article:||Ed Woolard: A life in governance.|