Synopsis and Objectives
In mid September 2005, Ashley Swenson, the chief financial officer (CFO) of a large computer-aided design and computer-aided manufacturing (CAD/CAM) equipment manufacturer needed to decide whether to pay out dividends to the firm’s shareholders, or to repurchase stock. If Swenson chose to pay out dividends, she would have to also decide upon the magnitude of the payout. A subsidiary question is whether the firm should embark on a campaign of corporate-image advertising, and change its corporate name to reflect its new outlook. The case serves as an omnibus review of the many practical aspects of the dividend and share buyback decisions, including (1) signaling effects, (2) clientele effects, and (3) the finance and investment implications of increasing dividend payouts and share repurchase decisions. This case can follow a treatment of the Miller-Modigliani1 dividend-irrelevance theorem and serves to highlight practical considerations to consider when setting a firm’s dividend policy. Suggested Questions for Advance Assignment to Students
The instructor could assign supplemental reading on dividend policy and share repurchases. Especially recommended are the Asquith and Mullins article2 on equity signaling, and articles by Stern Stewart on financial communication.3
1.In theory, to fund an increased dividend payout or a stock buyback, a firm might invest less, borrow more, or issue more stock. Which of those three elements is Gainesboro’s management willing to vary, and which elements remain fixed as a matter of the company’s policy? 2.What happens to Gainesboro’s financing need and unused debt capacity if: a. no dividends are paid?
b. a 20% payout is pursued?
c. a 40% payout is pursued?
d. a residual payout policy is pursued?
Note that case Exhibit 8 presents an estimate of the amount of borrowing needed. Assume that maximum debt capacity is, as a matter of policy, 40% of the book value of equity. 3. How might Gainesboro’s various providers of capital, such as its stockholders and creditors, react if Gainesboro declares a dividend in 2005? What are the arguments for and against the zero payout, 40% payout, and residual payout policies? What should Ashley Swenson recommend to the board of directors with regard to a long-term dividend payout policy for Gainesboro Machine Tools Corporation?
4. How might various providers of capital, such as stockholders and creditors, react if Gainesboro repurchased its shares? Should Gainesboro do so? 5.Should Swenson recommend the corporate-image advertising campaign and corporate name change to the Gainesboro’s directors? Do the advertising and name change have any bearing on the dividend policy or the stock repurchase policy that you propose?
Supporting Computer Spreadsheet Files
For students: Case_25.xls
For instructors: TN_25.xls
Hypothetical Teaching Plan
1.What are the problems here, and what do you recommend?
The CFO needs to resolve the issue of dividend payout in order to make a recommendation to the board. She must also decide whether to embark on a stock repurchase program given a recent drop in share prices. The problems entail setting dividend policy, deciding on a stock buyback, and resolving the corporate-image advertising campaign issue. But numerical analysis of the case shows that the problem includes other factors: setting policy within a financing constraint, signaling the directors’ outlook, and generally, positioning the firm’s shares in the equity market. 2.What are the implications of different payout levels for Gainesboro’s capital structure and unused debt capacity? The discussion here must present the financial implications of high-dividend payouts, particularly the consumption of unused debt capacity.
Because of the cyclicality of demand or overruns in investment spending, some attention might be given to a sensitivity analysis cast over the entire 2005 to 2011 period. 3.What is the nature of the dividend decision that Swenson must make? What are the pros and cons of the alternative positions? (Or alternatively, Why pay any dividends?) How will Gainesboro’s various providers of capital, such as its stockholders and bankers, react to a declaration of no dividend? What about the announcement of a 40% payout? How would they react to a residual payout? The instructor needs to elicit from the students the notions that the dividend-payout announcement may affect stock price and that at least some stockholders prefer dividends. Students should also mention the signaling and clientele considerations.
4.What risks does the firm face?
Discussion following this question should address the nature of the industry, the strategy of the firm, and the firm’s performance. This discussion will lay the groundwork for the review of strategic considerations that bears on the dividend decision. 5.What is the nature of the share repurchase decision that Swenson must make? How would this affect the dividend decision? The discussion here must present the repercussions of a share repurchase decision on the share price, as well as on the dividend question. Signaling and clientele considerations must also be considered.
6.Does the stock market appear to reward high-dividend payout? What about low-dividend payout? Does it matter what type of investor owns the shares? What is the impact on share price of dividend policy? The data can be interpreted to support either view. The point is to show that simple extrapolations from stock market data are untrustworthy, largely because of econometric problems associated with size and omitted variables (see the Black and Scholes article)
.4 7.What should Swenson recommend?
Students must synthesize a course of action from the many facts and considerations raised. The instructor may choose to stimulate the discussion by using an organizing framework such as FRICTO (flexibility, risk, income, control, timing, and other) on the dividend and share repurchase issues. The image advertising and name change issue will be recognized as another manifestation of the firm’s positioning in the capital markets, and the need to give effective signals.
The class discussion can end with the students voting on the alternatives, followed by a summary of key points. Exhibits TN1 and TN2 contain two short technical notes on dividend policy, which the instructor may either use as the foundation for closing comments or distribute directly to the students after the case discussion.
Gainesboro’s asset needs
The company’s investment spending and financing requirements are driven by ambitious growth goals (a 15% annual target is discussed in the case), which are to be achieved by a repositioning of the firm—away from its traditional tools-and-molds business and beyond its CAD/CAM business into a new line of products integrating hardware and software—to provide complete manufacturing systems. CAD/CAM commanded 45% of total sales ($340.5 million) in 2004 and is expected to grow to three-quarters of sales ($1,509.5 million) by 2011, which implies a 24% annual rate of growth in this business segment over the subsequent seven years.
In addition, international sales are expected to grow by 37% compounded over the subsequent seven years.5 By contrast, the presses-and-molds segment will grow at about 2.7% annually in nominal terms, which implies a negative real rate of growth in what constitutes the bulk of Gainesboro’s current business.6 In short, the company’s asset needs are driven primarily by a shift in the company’s strategic focus. Financial implications of payout alternatives
The instructor can guide the students through the financial implications of various dividend-payout levels either in abbreviated form (for one class period) or in detail (for two classes). The abbreviated approach uses the total cash flow figures (that is, for 2005–2011) found in the right-hand column of case Exhibit 8. In essence, the approach uses the basic sources-and-uses of funds identity:
Asset change = New debt + (Profits − Dividends)
With asset additions fixed largely by the firm’s competitive strategy, and with profits determined largely by the firm’s operating strategy and the environment, the remaining large-decision variables are changes in debt and dividend payout. Even additions to debt are constrained, however, by the firm’s maximum leverage target, a debt/equity ratio of 0.40. This framework can be spelled out for the students to help them envision the financial context.
Exhibit TN3 presents an analysis of the effect of payout on unused debt capacity based on the projection in case Exhibit 8. The top panel summarizes the firm’s investment program over the forecast period, as well as the financing provided by internal sources. The bottom panel summarizes the effect of higher payouts on the firm’s financing and unused debt capacity. The principal insight this analysis yields is that the firm’s unused debt capacity disappears rapidly, and maximum leverage is achieved as the payout increases. Going from a 20% to a 40% dividend payout (an increase in cash flow to shareholders of $95.6 million),7 the company consumes $134 million in unused debt capacity.
Evidently, a multiplier relationship exists between payout and unused debt capacity—every dollar of dividends paid consumes about $1.408 of debt capacity. The multiplier exists because a dollar must be borrowed to replace each dollar of equity paid out in dividends, and each dollar of equity lost sacrifices $0.40 of debt capacity that it would have otherwise carried.
Whereas the abbreviated approach to analyzing the implications of various dividend-payout levels considers total 2005 to 2011 cash flows, the detailed approach considers the pattern of the individual annual cash flows. Exhibit TN4 reveals that, although the debt/equity ratio associated with the 40% payout policy is well under the maximum of 40 in 2011, the maximum is breached in the preceding years. The graph suggests that a payout policy of 30% is about the maximum that does not breach the debt/equity maximum.
Exhibits TN5 and TN6 reveal some of the financial reporting and valuation implications of alternative dividend policies. Those exhibits use a simple dividend valuation approach and assume a terminal value estimated as a multiple of earnings. The analysis is unscientific, as the case does not contain the information with which to estimate a discount rate based on the capital asset pricing model (CAPM).9 The discounted cash flow (DCF) values show that the differences in firm values are not that large and that the dividend policy choice in this case has little effect on value. This conclusion is consistent with the Miller-Modigliani dividend-irrelevance theorem.
Regarding the financial-reporting effects of the policy choices, one sees that earnings per share (EPS on line 30 in Exhibits TN5 and TN6) and the implied stock price (line 31) grow more slowly at a 40% payout policy, because of the greater interest expense associated with higher leverage (see the cumulative source on line 22). Return on average equity (unused debt capacity on line 28) rises with higher leverage, however, as the equity base contracts. The instructor could use insights such as those to stimulate a discussion of the signaling consequences of the alternative policies, and whether investors even care about performance measures, such as EPS and return on equity (ROE).10
Neither the abbreviated nor detailed forecasts consider adverse deviations from the plan. Case Exhibit 8 assumes no cyclical downturn over the seven-year forecast period. Moreover, the model assumes that net margin doubles to 5% and then increases to 8%. The company may be able to rationalize those optimistic assumptions on the basis of its restructuring and the growth of the Artificial Workforce, but such a material discontinuity in the firm’s performance will warrant careful scrutiny. Moreover, continued growth may require new product development after 2006, which may incur significant research-and-development (R&D) expenses and reduce net margin.
Students will point out that, so far, the company’s restructuring strategy is associated with losses (in 2002 and 2004) rather than gains. Although restructuring appears to have been necessary, the credibility of the forecasts depends on the assessment of management’s ability to begin harvesting potential profits. Plainly, the Artificial Workforce has the competitive advantage at the moment, but the volatility of the firm’s performance in the current period is significant: The ratio of the cost of goods sold to sales rose from 61.5% in 2003 to 65.9% in 2004.
Meanwhile, the ratio of selling, general, and administrative expenses to sales is projected to fall from 30.5% in 2004 to 24.3% in 2005. Admittedly, the restructuring accounts for some of this volatility, but the case suggests several sources of volatility that are external to the company: economic recession, currency, new-competitor market entry, new product mishaps, cost overruns, and unexpected acquisition opportunities.
A brief survey of risks invites students to perform a sensitivity analysis of the firm’s debt/equity ratio under a reasonable downside scenario. Students should be encouraged to exercise the associated computer spreadsheet model, making modifications as they see fit. Exhibit TN7 presents a forecast of financial results, assuming a net margin that is smaller than the preceding forecasts by 1% and sales growth at 12% rather than 15%.
This exhibit also illustrates the implications of a residual dividend policy, which is to say the payment of a dividend only if the firm can afford it and if the payment will not cause the firm to violate its maximum debt ratios. The exhibit reveals that, in this adverse scenario, although a dividend payment would be made in 2005, none would be made in the two years that follow. Thereafter, the dividend payout would rise. The general insight remains that Gainesboro’s unused debt capacity is relatively fragile and easily exhausted.
The stock-buyback decision
The decision on whether to buy back stock should be that, if the intrinsic value of Gainesboro is greater than its current share price, the shares should be repurchased. The case does not provide the information needed to make free cash flow projections, but one can work around the problem by making some assumptions. The DCF calculation presented in Exhibit TN8 uses net income as a proxy for operating income,11 and assumes a weighted-average cost of capital (WACC) of 10%, and a terminal value growth factor of 3.5%. The equity value per share comes out to $35.22, representing a 59% premium over the current share price. Based on that calculation, Gainesboro should repurchase its shares.
Doing so, however, will not resolve Gainesboro’s dividend/financing problem. Buying back shares would further reduce the resources available for a dividend payout. Also, a stock buyback may be inconsistent with the message that Gainesboro is trying to convey, which is that it is a growth company. In a perfectly efficient market, it should not matter how investors got their money back (for example, through dividends or share repurchases), but in inefficient markets, the role of dividends and buybacks as signaling mechanisms cannot be disregarded. In Gainesboro’s case, we seem to have the case of an inefficient market; the case suggests that information asymmetries exist between company insiders and the stock market.
Clientele and signaling considerations
The profile of Gainesboro’s equity owners may influence the choice of dividend policy. Stephen Gaines, the board chair and scion of the founders’ families and management (who collectively own about 30% of the stock), seeks to maximize growth in the market value of the company’s stock over time. This goal invites students to analyze the impact of the dividend policy on valuation. Nevertheless, some students might point out that, as Gaines and Scarboro’s population of diverse and disinterested heirs grows, the demand for current income might rise. This naturally raises the question: Who owns the firm? The stockholder data in case Exhibit 4 show a marked drift over the past 10 years, moving away from long-term individual investors and toward short-term traders; and away from growth-oriented institutional investors and toward value investors.
At least a quarter of the firm’s shares are in the hands of investors who are looking for a turnaround in the not too distant future.12 This lends urgency to the dividend and signaling question. The case indicates that the board committed itself to resuming a dividend as early as possible —“ideally in the year 2005.” The board’s letter charges this dividend decision with some heavy signaling implications: because the board previously stated a desire to pay dividends, if it now declares no dividend, investors are bound to interpret the declaration as an indication of adversity. One is reminded of the story, “Silver Blaze,” written by Sir Arthur Conan Doyle featuring the famous protagonist Sherlock Holmes, in which Dr. Watson asks where to look for a clue:
“To the curious incident of the dog in the nighttime,” says Holmes. “The dog did nothing in the nighttime,” Watson answers.
“That was the curious incident,” remarked Sherlock Holmes.13
A failure to signal a recovery might have an adverse impact on share price. In this context, a dividend—almost any dividend—might indicate to investors that the firm is prospering more or less according to plan.
Astute students will observe that a subtler signaling problem occurs in the case: What kind of firm does Gainesboro want to signal that it is? Case Exhibit 6 shows that CAD/CAM equipment and software companies pay low or no dividends, in contrast to electrical machinery manufacturers, who pay out one-quarter to as much as half of their earnings. One can argue that, as a result of its restructuring, Gainesboro is making a transition from the latter to the former. If so, the issue then becomes how to tell investors.
The article by Asquith and Mullins14 suggests that the most credible signal about corporate prospects is cash, in the form of either dividends or capital gains. Until the Artificial Workforce product line begins to deliver significant flows of cash, the share price is not likely to respond significantly. In addition, any decline in cash flow, caused by the risks listed earlier, would worsen the anticipated gain in share price. By implication, the Asquith–Mullins work would cast doubt on corporate-image advertising. If cash dividends are what matters, then spending on advertising and a name change might be wasted.
Stock prices and dividends
Some of the advocates of the high-dividend payout suggest that high stock prices are associated with high payouts. Students may attempt to prove that point by abstracting from the evidence in case Exhibits 6 and 7. As we know from academic research (for example, Friend and Puckett),15 proving the relationship of stock prices to dividend payouts in a scientific way is extremely difficult. In simpler terms, the reason is because the price/earnings (P/E) ratios are probably associated with many factors that may be represented by dividend payout in a regression model. The most important of those factors is the firm’s investment strategy; Miller and Modigliani’s16 dividend-irrelevance theorem makes the point that the firm’s investments—not the dividends it pays—determine the stock prices.
One can just as easily derive evidence of this assertion from case Exhibit 7. The sample of zero-payout companies has a higher average expected return on capital (24.9%) than the sample of high-payout companies (average expected return of 9.4%); one may conclude that zero-payout companies have higher returns than the high-payout companies and that investors would rather reinvest in zero-payout companies than receive a cash payout and be forced to redeploy the capital to lower-yielding investments.
The decision for students is whether Gainesboro should buy back stock or declare a dividend in the third quarter (although, for practical purposes, students will find themselves deciding for all of 2005). As the analysis so far suggests, the case draws students into a tug-of-war between financial considerations, which tend to reject dividends and buybacks at least in the near term, and signaling considerations, which call for the resumption of dividends at some level, however, small. Students will tend to cluster around the three proposed policies: (1) zero payout, (2) low payout (1% to 10%), and (3) a residual payout scheme calling for dividends when cash is available.
The arguments in favor of zero payout are: (1) the firm is making the transition into the CAD/CAM industry, where zero payout is the mode; (2) the company should not ignore the financial statements and act like a blue-chip firm—Gainesboro’s risks are large enough without compounding them by disgorging cash; and (3) the signaling damage already occurred when the directors suspended the dividend in 2005.
The arguments in favor of a low payout are usually based on optimism about the firm’s prospects and on beliefs that Gainesboro has sufficient debt capacity, that Gainesboro is not exactly a CAD/CAM firm, and that any dividend that does not restrict growth will enhance share prices. Usually, the signaling argument is most significant for the proponents of this policy. The residual policy is a convenient alternative, although it resolves none of the thorny policy issues in the case. A residual dividend policy is bound to create significant signaling problems as the firm’s dividend waxes and wanes through each economic cycle.
The question of the image advertising and corporate name change will entice the naive student as a relatively cheap solution to the signaling problem. The instructor should challenge such thinking. Signaling research suggests that effective signals are both unambiguous and costly. The advertising and name change, costly as they may be, hardly qualify as unambiguous. On the other hand, seasoned investor relations professionals believe that advertising and name changes can be effective in alerting the capital markets to major corporate changes when integrated with other signaling devices such as dividends, capital structure, and investment announcements. The whole point of such campaigns should be to gain the attention of the “lead steer” opinion leaders.
Overall, inexperienced students tend to dismiss the signaling considerations in this case quite readily. On the other hand, senior executives and seasoned financial executives view signaling quite seriously. If the class votes to buy back stock or to declare no dividend in 2005, asking some of the students to dictate a letter to shareholders explaining the board’s decision may be useful. The difficult issues of credibility will emerge in class with a critique of this letter.
If the class does vote to declare a dividend payout, the instructor can challenge the students to identify the operating policies they gambled on to make their decision. The underlying question: If adversity strikes, what will the class sacrifice first: debt, or dividend policies?
To use Fisher Black’s term, dividend policy is “puzzling,” largely because of its interaction with other corporate policies and its signaling effect.17 Decisions about the firm’s dividend policy may be the best way to illustrate the importance of managers’ judgments in corporate finance. However the class votes, one of the teaching points is that managers are paid to make difficult, even high-stakes policy choices on the basis of incomplete information and uncertain prospects. Exhibit TN1
GAINESBORO MACHINE TOOLS CORPORATION
The Dividend Decision and Financing Policy
The dividend decision is necessarily part of the financing policy of the firm. The dividend payout chosen may affect the creditworthiness of the firm and hence the costs of debt and equity; if the cost of capital changes, so may the value of the firm. Unfortunately, one cannot determine whether the change in value will be positive or negative without knowing more about the optimality of the firm’s debt policy. The link between debt and dividend policies has received little attention in academic circles, largely because of its complexity, but it remains an important issue for chief financial officers and their advisors. The Gainesboro case illustrates the impact of dividend payout on creditworthiness.
Dividend payout has an unusual multiplier effect on financial reserves. Table TN1 varies the total 2005–2011 sources-and-uses of funds information given in case Exhibit 8, according to different dividend-payout levels. Exhibit TN1 (continued)
Exhibit TN1 (continued)
As Table TN1 reveals, one dollar of dividends paid consumes $1.40 in unused debt capacity. At first glance, this result seems surprising—under the sources-and-uses framework, one dollar of dividend is financed with only one dollar of borrowing. The sources-and-uses reasoning, however, ignores the erosion in the equity base: A dollar paid out of equity also eliminates $0.40 of debt that the dollar could have carried. Thus, a multiplier effect exists between dividends and unused debt capacity, whenever a firm borrows to pay dividends.
Choosing a dividend payout will affect the probability that the firm will breach its maximum target leverage. Figure TN1 traces the debt/equity ratios associated with Gainesboro’s dividend-payout ratios.
Plainly, the 40% dividend-payout ratio violates Gainesboro’s maximum debt/equity ratio of 40%.
The conclusion is that, because the dividend policy affects the firm’s creditworthiness, senior managers should weigh the financial side effects of their payout decisions, along with the signaling, segmentation, and investment effects, to arrive at their final decision for the dividend policy. Exhibit TN2
GAINESBORO MACHINE TOOLS CORPORATION
Setting Debt and Dividend-Payout Targets
The Gainesboro Machine Tools Corporation case well illustrates the challenge of setting the two most obvious components of financial policy: target payout and debt capitalization. The policies are linked with the firm’s growth target, as shown in the self-sustainable growth model:
gss = (P/S × S/A × A/E)(1 − DPO)
gss is the self-sustainable growth rate
P is net income
S is sales
A is assets
E is equity
DPO is the dividend-payout ratio
This model describes the rate at which a firm can grow if it issues no new shares of common stock, which describes the behavior or circumstances of virtually all firms. The model illustrates that the financial policies of a firm are a closed system: Growth rate, dividend payout, and debt targets are interdependent. The model offers the key insight that no financial policy can be set without reference to the others. As Gainesboro shows, a high dividend payout affects the firm’s ability to achieve growth and capitalization targets and vice versa. Myopic policy—failing to manage the link among the financial targets—will result in the failure to meet financial targets.
Setting Debt-Capitalization Targets
Finance theory is split on whether gains are created by optimizing the mix of debt and equity of the firm. Practitioners and many academicians, however, believe that debt optima exist and devote great effort to choosing the firm’s debt-capitalization targets. Several classic competing considerations influence the choice of debt targets:
1.Exploit debt-tax shields. Modigliani and Miller’s theorem implies that in the world of taxes, debt financing creates value.1 Later, Miller theorized that when personal taxes are accounted for, the leverage choices of the firm might not create value. So far, the bulk of the empirical evidence suggests that leverage choices do affect value. 2.Reduce costs of financial distress and bankruptcy. Modigliani and Miller’s theory naively implied that firms should lever up to 99% of capital. Virtually no firms do this. Beyond some prudent level of debt, the cost of capital becomes very high because investors recognize that the firm has a greater probability of suffering financial distress and bankruptcy. The critical question then becomes: What is “prudent”? In practice, two classic benchmarks are used: a. Industry-average debt/capital: Many firms lever to the degree practiced by peers, but this policy is not very sensible. Industry averages ignore differences in accounting policies, strategies, and earnings outlooks. Ideally, prudence is defined in firm-specific terms.
In addition, capitalization ratios ignore the crucial fact that a firm goes bankrupt because it runs out of cash, not because it has a high debt/capital ratio. b. Firm-specific debt service: More firms are setting debt targets based on the forecasted ability to cover principal and interest payments with earnings before interest and taxes (EBIT). This practice requires forecasting the annual probability distribution of EBIT and setting the debt-capitalization level, so that the probability of covering debt service is consistent with management’s strategy and risk tolerance. 3.Maintain a reserve against unforeseen adversities or opportunities. Many firms keep their cash balances and lines of unused bank credit larger than may seem necessary, because managers want to be able to respond to sudden demands on the firm’s financial resources caused, for example, by a price war, a large product recall, or an opportunity to buy the toughest competitor.
Academicians have no scientific advice about how large those reserves should be. 4.Maintain future access to capital. In difficult economic times, less creditworthy borrowers may be shut out from the capital markets and, thus, unable to obtain funds. In the United States, “less creditworthy” refers to the companies whose debt ratings are less than investment grade (which is to say, less than BBB2 or Baa3). Accordingly, many firms set debt targets in such a way as to at least maintain a creditworthy (or investment grade) debt rating. 5.Opportunistically exploit capital-market windows. Some firms’ debt policies vary across the capital-market cycle. Those firms issue debt when interest rates are low (and issue stock when stock prices are high); they are bargain-hunters (even though no bargains exist in an efficient market). Opportunism does not explain how firms set targets so much as why firms deviate from those targets.