Vanderbilt university roundtable on the capital structure puzzle

Abstract

April 2, 1998         Nashville, Tennessee

JOEL STERN: Good afternoon. I'm Joel Stern, managing partner of Stern Stewart & Co., and, on behalf of our hosts here at Vanderbilt's Owen Graduate School of Management, I want to welcome you all to this discussion of corporate capital structure. Before getting into our subject matter, let me take a moment to thank Hans Stoll for organizing this conference on “Financial Markets and the Corporation.” I also want to take this opportunity to salute Professor Martin Weingartner—in whose honor this conference is being held—at the conclusion of a long and productive career. Marty's contributions to the field of corporate finance are many and considerable; and, though he may be stepping down from his formal position, we expect to continue to hear from him for many more years.

The subject of today's meeting is corporate capital structure: Does capital structure matter? And, if so, how and why does it matter? Although these questions have been seriously debated in the academic finance profession for almost 40 years, we seem to be no closer to a definitive answer than we were in 1958, when Merton Miller and Franco Modigliani published their article presenting the first of their two famous “irrelevance” propositions.

Following the M&M propositions, academic researchers in the 1960s and 1970s turned their attention to various market “imperfections” that might make firm value depend on capital structure and dividend policy. The main suspects were (1) a tax code that encourages debt by making interest payments, but not dividends, tax deductible, and (2) expected costs of financial distress, including corporate underinvestment, that can become important as you increase the amount of debt in the capital structure. Toward the end of the 1970s, there was also discussion of “signaling” effects—for example, the tendency for the stock market to respond negatively to announcements of new stock issues.

A defining moment in the academic capital structure debate came in 1984, when Professor Stewart Myers devoted his Presidential address to the American Finance Association to something he called “The Capital Structure Puzzle.” The puzzle was this: Most academic discussions of capital structure were based on the assumption that companies make financing decisions that are guided by a target capital structure—a proportion of debt to equity that management aims to achieve, if not at all times, then at least as a long-run average. But the empirical evidence suggested otherwise. Rather than adhering to targets, Professor Myers observed, most large U.S. public companies behaved as if they were following a financial “pecking order.” They were funding investment with retained earnings rather than external financing if possible; and if external funding was necessary, they issued debt first and equity only as a last resort.

Since then, the capital structure debate has raged on. Harvard professor Michael Jensen entered it in the mid-1980s, pointing to the success of LBOs and citing the beneficial effect of debt financing on management's tendency to overinvest in industries with excess capital and capacity. And, as if to oblige Professor Jensen, the market continued to supply large numbers of LBOs and other highly leveraged transactions throughout the rest of the decade. Then, in the early 1990s, we saw an almost complete halt to leveraged deals. But today, of course, leverage is back. A new wave of LBOs has shattered most of the old records, and junk-bond issuance is at all-time highs.

So, if capital structure is irrelevant, then what's going on here? And do the successes of the LBO movement have anything to say to the managements of our largest public companies? Such effects seemed to confirm the existence of large “information costs” that might also influence corporate financing choices in predictable ways.

With us here to discuss these issues, and to help us shed some light on this capital structure puzzle, is a small, but distinguished group of academics and practitioners. And let me introduce them briefly:

STEWART MYERS, whose name I have already mentioned several times, is the Billard Professor of Corporate Finance at one of my favorite schools, the MIT Sloan School of Management. Stew has done research over the years in issues dealing with capital structure, valuation, and regulation. He has also been an adviser to a large number of corporations and financial institutions. And let me mention that, for us at Stern Stewart, the name Stewart Myers has a special significance. As most people here are well aware, Stew is the co-author, with Dick Brealey of the London Business School, of Principles of Corporate Finance, the leading textbook in corporate finance. Every person who gets hired at Stern Stewart is required to have read that volume by the time he or she walks in the door.

ALICE PETERSON is Vice President and Treasurer of Sears, Roebuck and Co. As we all know, Sears has made dramatic improvements in operating performance and achieved large increases in shareholder value over the past few years. Alice joined Sears in 1989 as Director of Corporate Finance, 4 years before being made Vice President and Treasurer in 1993. Prior to coming to Sears, she held corporate finance and treasury positions at Kraft and at Pepsico. She earned her MBA here at Vanderbilt in 1981, and she serves on the boards of two New York Stock Exchange companies.

JOHN McCONNELL is the Emanuel T. Weiler Professor of Finance at Purdue University. John has made extensive contributions to the field of corporate finance, especially in the analysis of innovative securities. I still remember an article on income bonds that John contributed to our old Chase Financial Quarterly, the original predecessor of our Journal of Applied Corporate Finance. And, in an issue of the JACF about 5 or 6 years ago, John coauthored (with Eduardo Schwartz) a fascinating account of the origins of LYONS, the very successful puttable, convertible securities pioneered by Merrill Lynch. Given John's ability to tell stories, none of us is surprised that he manages to win teaching awards at Purdue year after year.

Last is DENNIS SOTER, my colleague and fellow partner at Stern Stewart. After graduating from the University of Rochester's Simon School of Business in 1972, Dennis joined me at the Chase Manhattan Bank. Then, in 1979, we parted ways. Dennis became Vice President in charge of corporate development at Brown Foreman, where he helped transform the firm from the maker of Jack Daniels (and other mild intoxicants) into a diversified consumer goods firm. Then he went to Ernst & Whinney, where he was the national practice director of M&A. Next, he joined Citizens Utilities and helped them to become something of an unregulated company as well as a regulated company.

Several years ago, we were very fortunate in persuading each other that we should be together again. And Dennis now runs our corporate finance advisory activity and also oversees implementations of EVA in middle market-sized companies. In the past 2 years, as I'm sure Dennis will tell us, he served as financial adviser in three highly successful leveraged recapitalizations. Each of these three deals involved borrowing substantial amounts of new debt to buy back shares—and two involved major changes major changes in dividend policy as well.

STERN: So, now that we know who the panelists are, let me turn the floor over to Professor Stewart Myers. You might not remember this, Stew, but in 1983 you were kind enough to publish an article in our Midland Corporate Finance Journal entitled “The Search for Optimal Capital Structure.”

And it was not only a marvelous piece, but it was the lead article in the very first issue—Volume 1 Number 1—of that journal. Then, about 10 years passed, and you wrote a second article for us that was called “Still Searching for Optimal Capital Structure.”

And that was also a wonderful piece—one that I use (along with the first one) in the courses I teach at Columbia and Carnegie Mellon.

My question for you is: What is your current thinking on the capital structure debate? And what are you going to call your next article?

STEWART MYERS: The next one is going to be called “Stop Searching for Optimal Capital Structure.” Let me take a minute or two to tell you why. Optimal capital structure is obviously something I've been concerned and struggling with ever since I arrived at the doctoral program at Stanford. I've been frustrated over the years by our inability to come up with any simple answer. But I am lately coming to a different view. Maybe it will start the discussion off.

When we talk about optimal capital structure, we are thinking of the percentages of different securities on the right-hand side of the balance sheet. We tend to think in terms of the mix of debt and equity, at least to start, and then go on to consider other securities as well. We look at preferred stock, for example, and at hybrid securities like convertible debt.

At what level do we understand how these financing choices affect firm value? At a tactical level, I think we understand it very well. For example, if you ask either academics or practitioners to analyze a financial innovation like the tax-deductible preferreds that John McConnell just finished telling us about, we do that pretty well. We understand how those things work, why they're designed the way they are, and what you need to do to get them sold. So, at this tactical level, we can be fairly satisfied with our understanding of capital structure.

Where we tend to fall down in terms of neat or simple theories is in understanding the role of capital structure at what I will call the “strategic” level—that is, when you're trying to explain the debt ratios of companies on average or over long periods of time. We do have some useful insights about capital structure, and we know what ought to matter. But it's very difficult to put these insights together into a simple theory that predicts what managers are going to do—or tells us what they should be doing.

Why aren't we cracking the problem? I think we are starting in the wrong place. We shouldn't be starting with the percentages of different kinds of financing on the right-hand side of the balance sheet. We should not be starting with capital structure, but with financial structure. By financial structure I mean the allocation of ownership and control, which includes the division of the risk and returns of the enterprise—and particularly of its intangible assets—between the insiders in the firm and the outsiders.

What's really going on in the public corporation is a coinvestment by insiders, who bring their human capital to it, and by financial investors. These two groups share the intangible assets of the organization. And, in order to make that shared investment work, you've got to figure out issues of ownership and control, incentives, risk-sharing, and so forth. You've got to start by making sure you get all of those things right. That's what I call financial structure.

Financial structure is not the same thing as a financing mix. Let me offer a simple example. Take a high-tech venture capital startup and compare it to Microsoft. Even though their debt ratios are likely to be identical—that is, zero—I think we'd all agree that they are not the same thing from a financial point of view. The same comparison could be made between a publicly held management consulting company and the usual private management consulting firm. Though their balance sheets might look exactly the same, I would say they're very different. In particular, I would predict that if you take a small consulting company public, it's almost sure to crater. The assets go home every night, and it's going to be very difficult to have the right incentives to keep key people in the company while at the same time satisfying outside investors.

So, if Stern Stewart ever goes public, Joel, I'm going to wait 6 months and then sell you short.

STERN: I actually had a dream that somebody was going to do that to us—but I had no idea, Stew, that it was going to be you! I must confess that such thoughts have passed through my mind. But, if we were to take such a course (which we have no plan to do), we would design the new firm so as to protect outside investors not so much from the possibility that the assets go home at night, but that they might move to the island of Maui, permanently.

Incidentally, this is not an unreasonable issue because Booz-Allen did that once. You may not be aware of it, but in the 1970s Booz-Allen went public at a price in the mid-teens. The shares went as high as about $20 and then dropped to about $2—and they stayed between $2 and $6 for quite a while. The reason I know about it is that I served on the board of directors of a company with the Vice Chairman of Booz-Allen. And every time the possibility of going public was raised for this privately held company, he would say: “It's the wrong thing to do.” And he would drag us through this horrible experience once more.

But Booz-Allen made a fundamental mistake—one that, if corrected, might have changed the outcome. They didn't differentiate between outsider shares and insider shares. If the insiders own shares that only gradually convert to outsider shares, the insiders aren't going anywhere. They could also have issued stock options to the employees with values tied to the value of the insider shares. With insider shares and options, it would take a long time for people to take out the wealth that they were building up in the firm.

How do you feel about that idea, Stew?

MYERS: I think you're just making my point. First of all, I don't think that it makes sense to take a small consulting company public. But if you did, you'd have to pay attention to financial structure—to issues like who owns what, who gets to make what decisions, what are the goals and performance measures, and how are people rewarded for meeting them.

STERN: So, your concept of financial structure is essentially the same as what some people are calling corporate governance or organizational structure—it's the assignment of decision rights, monitoring, performance measures, incentives, and all of that?

MYERS: Financial structure is my shorthand—just two words—for all of that sort of stuff. My point is that these issues of financial structure are the first-order concerns of most corporations. Management has to get them right.

Now, that's not the same as saying that managers have to do all these things consciously. Sometimes they get it right because they do what other corporations have done and survived.

Once an industrial corporation goes public, it's ordinarily a mid-cap firm with a financial structure well in place. In a sense, it has solved 80% of its problem—that is, as long as it keeps operating efficiently. Having solved that problem, I don't think most corporations worry a great deal about their debt-equity ratio. Obviously, they worry if it gets too high and they worry if it gets too low. But there's a big middle range where it doesn't seem to matter very much. I think that's why we're having a hard time pinning down exactly what the debt-equity ratio is or should be. It's a “second order” thing compared to the choice of financial structure.

STERN: Well, people have pointed out that the tax-deductibility of interest expense makes debt a way of adding considerable value. And because of the value added by those tax shields, one could argue that aggressive use of debt may turn out to be the best defense against an unfriendly takeover. Doesn't that argument seem to imply that capital structure could at least become a first-order concern?

MYERS: By raising the issue of hostile takeovers, I think you are mixing up the investment decision with the financing decision. I would like to keep them separate, at least as a starting point. Most takeover targets become targets not because they don't have enough debt, but because of bad investment decisions and poor operating performance. The primary cause of the wave of LBOs in the 1980s was not corporate failure to exploit the tax advantage of debt. Most of the LBOs—and I'm oversimplifying a little—were “diet deals.” They involved taking over mature companies with too much cash and too few investment opportunities and putting them on a diet. So, although there were tax savings, the transactions themselves were not tax-driven.

DENNIS SOTER: Most of the companies that we have worked with over the years do spend a lot of time thinking about capital structure. Now, it's true that they typically do not think about it the way academics do—that is, in terms of a market debt-to-equity ratio. Most of them define their objectives in terms of bond ratings or book leverage. But they do have loosely defined financial strategies and some idea of what constitutes an appropriate mix of debt and equity in their capital structure.

I'd like to ask Professor Myers a question: You used the term “optimal capital structure.” Would you define that for us?

MYERS: In its simplest terms, optimal capital structure is the optimal percentage of debt on the balance sheet.

SOTER: But optimal from what standpoint?

MYERS: Maximizing the market value of the company. I'm not saying that's the wrong question, or an irrelevant or an uninteresting question. It's a very interesting question. But I think by focusing too narrowly on that, you miss these other things which I put under the rubric of financial structure—things which I tend to think are more important to corporate managers, and so more powerful in explaining the corporate behavior we see.

STERN: I think you're right, at least in the sense that I regularly see behavior that doesn't appear to me to be value-maximizing. For example, in a roundtable discussion we ran a few years ago, one of the participants was an owner of a publicly traded company where he and his family owned about 37% of the equity. And I asked him why his debt ratio was so low. I said to him, “You're volunteering to pay an awful lot in taxes that you could avoid. Why don't you raise your debt ratio to the point where most of your pre-tax operating income is tax-sheltered?”

You see, the amount of taxes he was paying was quite sizeable; it was some $12–$15 million a year that was literally going out the door that could have been saved just by changing the capital structure. Why do companies do that, Stew?

MYERS: Tell me what he said first.

STERN: He looked at me somewhat quizzically, and he said, “Well, don't you have to pay taxes?”

MYERS: You're absolutely right. If you look at taxes alone, and you calculate the present value of the taxes that could be saved by greater use of debt, it seems like a big number. It's hard to explain why corporations don't seem to work very hard to take advantage of the tax shelter afforded by debt. I can't really explain why companies aren't taking advantage of those savings.

I also agree with Dennis that corporations will say that they have target debt ratios. But the fact is they don't work very hard to get there. If a company is very profitable and doesn't have a need for external funds, it's probably going to work down to a low debt ratio. If it's short of funds, it's going to work up to a high debt ratio. These companies are not treating their debt ratio targets as if they were first-order goals.

SOTER: Let me offer one possible clue to this capital structure puzzle by asking a question: Is it in the personal interests of the chief executive officer and the chief financial officer to employ leverage aggressively? Are they motivated through incentives, either through stock ownership or otherwise, to have an aggressive debt policy?

I submit that very few CFOs of the largest U.S. public companies have enough equity ownership to even consider undertaking a leveraged recapitalization. If it's successful, he or she looks good for the moment. But if it's doesn't work, he'll never get another job in corporate America as a chief financial officer. So there is a great deal of personal risk for corporate management for which there may be little compensating reward. Without a significant ownership interest, who wouldn't prefer to have a single-A bond rating and sleep well at night?

STERN: Unless they were subjected to an unfriendly takeover—in which case they would lose their jobs for sure.

SOTER: Joel, you're citing an extreme example, one where there's so much unused debt capacity that a company invites a hostile bid. In my experience, there are many other companies that, although not takeover candidates, still have enough excess debt capacity that their shareholders would benefit from a leveraged recapitalization.

JOHN McCONNELL: When I started my career many years ago, I taught a course on corporate capital structure, thinking that I knew something about it. Then, for a period of a few years, I quit teaching capital structure because I concluded I knew absolutely nothing. More recently, I've started teaching a course that I call capital structure, but which really amounts to a course in corporate governance. And, so, I think I've been undergoing an evolution in thinking that is quite similar to the one Stew has just described.

And like you, Dennis, as Stew was defining optimal capital structure and describing the capital structure puzzle, I too was thinking about management incentives. Most of us, I suspect, would agree with the general proposition that there is a tax shield associated with debt financing. There would certainly be some disagreement about how valuable that tax shield is: Does it amount to a full 34 cents on the dollar of debt financing, or does that number fall to only 15 or 20 cents when you consider the taxes paid by debt holders? But, regardless of which end of this range you choose, most people in this room would probably agree that the value of such savings can be substantial. And thus most of us here would also likely agree that, for whatever reason, many public corporations are not using enough debt—not using the value-maximizing level of debt.

But let me also respond to Dennis by saying that stock ownership may not be a complete answer to the question. I'm on the board of directors of a bank with about $600 million in assets, and there is great latitude as to the minimum capital ratio that the bank can have. The owners of the bank, however, insist for some reason upon keeping their capital ratio high. In so doing, they forgo not only possible tax benefits, but also the deposit insurance “funding arbitrage.”

Like the company Joel was describing earlier, 47% of the stock of this particular bank is owned by a single family of investors. And if one thinks only in terms of their ownership incentives, it would seem that that particular ownership structure would be one that would have sufficiently strong incentives to induce managers to have a high leverage ratio. But instead, they have low leverage and considerable excess capital.

And this case, by the way, is by no means an anomaly; it is quite representative of U.S. public corporations with heavy insider ownership. The studies that have looked at the relationship between insider stock ownership and leverage ratios show that, beyond a certain point, companies with very large insider ownership tend to have lower-than-average leverage ratios. Like other academic studies in which the relationship between ownership concentration and stock value is shown to be a bell-shaped curve, these studies of leverage and insider ownership suggest that insiders can actually own too much stock in their own firm.

STERN: I was thinking along the same lines, John, when we were doing some work with the Coca-Cola Company. The CEO of Coke had a very large percentage of his own personal net worth tied up in the equity of the company. He was the third or fourth largest shareholder. And it occurred to me that the concentration of the CEO's wealth in Coke stock, and thus in an undiversified portfolio, could be a major factor in keeping the company from having a high debt ratio. That is, given his personal portfolio, it may have been rational for him to keep the financial risk of the company to a minimum.

What this suggests to me is that, in cases with very large inside owners, we may need to devise a compensation scheme that helps overcome the owners’ risk aversion so as to align their risk-reward tradeoff with that of well-diversified shareholders.

McCONNELL: What about the use of nonvoting stock? That would allow insiders to raise capital for investment opportunities without reducing their control over decision-making. For example, the CEO of Coca-Cola could have had the company issue nonvoting stock, and then concentrated his own holdings in the voting stock. Is something like that what you have in mind?

STERN: No, that wouldn't be a solution to the problem I'm thinking about. The concern I have is that because the CEO's holdings represent such a large percentage of his own personal net worth, he was much more risk averse than institutional investors with well diversified portfolios. And this risk aversion may well have caused him to use less than the amount of leverage that would have optimal from the perspective of his institutional owners.

McCONNELL: One possible solution to this problem—not for Coke, but at least in the case of small, closely held companies—would be for the CEO to sell a large fraction of his equity, thus allowing him to diversify his portfolio and allowing outsiders to have greater control over the firm. This could end up significantly increasing the value of the firm, at least in countries like the U.S. with its strong legal protections for small investors and well-functioning corporate control market.

STERN: But, to return to our subject, we really don't have an explanation why public companies have low debt ratios—or at least low enough that they end up paying millions of dollars more in taxes than seems necessary. Since we have with us a senior executive from one of America's largest public companies, why don't we turn to her? Alice, what's your explanation of all this? And how do you think about corporate financial policy at Sears?

ALICE PETERSON: Let me start by reinforcing Stewart Myers's point that capital structure is a relatively small part of a much larger, value-creating equation. At Sears, as in most companies, creating shareholder value is the main governing objective. Behind our mantra to make Sears a compelling place to shop, work, and invest, we use an EVA-type metric to track performance in the “invest” category.

In order to create value for investors, companies need a business model to guide them in generating excess returns. Business models vary widely from company to company, even within the same industry. Such a business model incorporates overall objectives, organization, strategies, and financial goals. It occurs to me that Stewart Myers's broad definition of capital structure runs through our entire business model.

We call our overall financial metric Shareholder Value Added. It is simply operating profits less the cost of all the capital it takes to create those profits. When we break down capital costs into the amount of capital times the cost of capital, we focus significantly more managerial effort on the amount of capital—and our approach is operationally focused. In other words, we want to attack the root need for the capital—and, for us, that's inventory and stores and fixturing. I often say that, of our 350,000 associates at Sears, nearly all of us have a daily impact on the amount of capital, but only a handful decide how to fund the capital needs.

So, we spend disproportionately more time talking about how to get more profit using less assets. For the few of us who do focus on the righthand side of the balance sheet, there is considerable effort aimed at achieving the lowest long-term cost of capital by managing the capital structure.

How does a company decide its optimal mix of debt and equity? At Sears we start by asking how much financial flexibility we want to pay for. That leads to a discussion in which we determine our target debt rating, which in turn translates into a target capital structure. With every strategic plan we're gauging the extent of free cash flow, the appropriate level of capital expenditures, and whether and to what extent we should be buying back stock. I can tell you we have dug seriously into the capital structure question several times over the last 5 years as we have IPO'd and spun off various business units and not only for the businesses leaving the portfolio, but for those remaining, too.

So, what are these financial flexibility considerations? They include growth prospects and whether a company is inclined to grow organically or by acquisitions. They also include the need to position the company to weather extreme economic cycles and exogenous risks. Other considerations are ownership structure and stock positioning, as well as dividend policy, and our ability to manage overall enterprise risk.

One consideration that doesn't show up on most companies’ list, but which is very important for Sears, is its ongoing debt-financing needs. Sears is different from its retail competitors by virtue of our successful credit business. Our $28 billion consumer receivables portfolio is the largest proprietary retail credit portfolio by a wide margin (the next largest is J.C. Penney at $5 billion). The $25+ billion in debt financing required to fund this profitable credit business is a key consideration in determining our target debt rating. We have targeted a ‘Single-A’ long-term debt rating to optimize our overall cost of capital. We think a “Double-A” rating is entirely too expensive, but feel that “Triple-B” greatly limits our flexibility. Importantly, impacts from a downgrade would disproportionately affect Sears versus our retail and service competition.

To give you a sense of how we incorporate this shareholder value discipline into our daily lives, the way we approach target-setting for the enterprise-wide capital structure is by solving for the capital structure that optimizes our market value. We take into consideration historical performance, economic scenarios, business and industry prospects, risk of bankruptcy, and capital constraints. Coming at it from the other side, we also focus on our individual businesses’ cost of capital. Our individual businesses include very different formats: full-line stores, Homelife furniture stores, Sears Hardware stores, our services business, and our credit card business, to name just a few. These businesses represent different levels of risk, and so warrant different hurdle rates.

We address the individual businesses’ cost of capital in two ways: First, we take a rating agency view of the business. We ask ourselves, “What are the business's prospects for cashflow, and what are the associated risks?” We also assess the business's competitive position and develop an industry outlook. Second, we approach the same business's cost of capital by looking at its competitors and other benchmarks. We ask: “What is their capital structure, and their cost of capital? And are there tax issues?” We want to know whether our financial “DNA” is a help or a hindrance to competing effectively. We look at on versus off-book capital. We do analytics, we compare managements, and we look at “the story” behind each business. Needless to say, it's important that the cost of capital and the capital structure for all our component businesses add up to the enterprise-wide picture. By pursuing this simultaneous top-down and bottom-up approach, we find we are getting at all the issues that allow us to get more debt in our capital structure for the same amount of financial flexibility.

STERN: Well, to return to my earlier question, are there many public companies that are passing up opportunities to add significant value by raising their debt ratios? Or are we missing something important here?

McCONNELL: Well, Joel, I think we've already heard the best answer we're going to get. Our incentive structures are not very effective in overcoming managers’ risk aversion; there just isn't enough reward for success to justify the personal risks to management.

STERN: But if the incentive structures are not effective, then why don't we see more unfriendly takeovers to make sure that they are aligned properly?

PETERSON: Well, let me remind you both of your earlier comments. You both told stories in which owners with a long-term commitment to an institution were reluctant to jeopardize the future of that institution by taking on too much debt, or operating with insufficient capital. There are costs—significant potential costs—to operating a business with too much leverage, or too little equity.

When I think about what gets written up about public companies—more so by fixed income analysts and rating agencies than by equity analysts—capital structure is very much talked about. It becomes the issue—and for equity as well as fixed income analysts—when a heavy debt load significantly reduces a company's flexibility. And if for no other reason than this, companies are very aware of it. I know that we focus very much on our competitors’ capital structures. We often consider what their balance sheets permit them to do in the way of competitive strikes. To listen to the academics here, we'd believe that corporations don't proactively engineer their capital structure. And to listen to Joel and Dennis, we'd believe few companies in America have enough debt. I think both notions are unfounded.

So, I must confess that this idea that public companies are systematically underleveraged is Greek to me. Within our current business model, our company is not underleveraged today. We were clearly overleveraged 5 years ago.

SOTER: I want to respond to this issue of financial flexibility, and I want to do so by elaborating on my earlier statement about management's reluctance to make aggressive use of leverage. There's more to my explanation than just risk aversion. And it comes down to this: Corporate managers don't like to have their strategies subjected to the scrutiny of markets. They like to be able to draw down bank lines and write a check. All things equal, they would prefer not to have to act like an LBO firm in which every deal is funded on a one-off basis.

I would take issue with Alice's argument that most public companies are not underleveraged. I do not think that a “Single-A” rating is likely to be a value-maximizing capital structure for most companies. We have looked at about 25 different industries; and, with few exceptions, our analysis suggests that the optimal, value-maximizing capital structure is below investment grade.

McCONNELL: How do you know that, though?

SOTER: Without going into details, our method involves an assessment of the trade-off between realizing the tax benefits of debt financing versus the increased probability of financial distress and its associated costs.

PETERSON: Well, that's all well and good. But how would I get $25 billion on an ongoing basis with a sub-investment-grade rating?

SOTER: You could issue subordinated debt.

PETERSON: Dennis, the entire high-yield debt market in recent years was only $40 billion. And costs can be considerably above investment-grade debt costs.

SOTER: I agree that the costs of the debt will go up. As we all know, the cost of debt is always going to go up as you use more of it. But I'm looking at debt as a substitute for higher-cost equity. And the question I'm trying to answer is this: How much debt can you issue until your debt and equity become so risky that your weighted average cost of capital begins to go up. In other words, at what level do you minimize capital costs?

What I am suggesting is that most public companies would actually reduce their weighted average cost of capital and increase their overall value by using debt to the point where their debt rating falls below investment grade—not far below investment grade, but below investment grade.

PETERSON: Yes, but that would significantly reduce the financial flexibility of the company, and potentially destroy more value than it creates. Your financing strategy might work for companies without the scope and aspirations of, say, the company I work for. But when you want to grow, either internally or by acquisition, and when you have a $25 billion financing requirement, you're going to want the financial flexibility that comes with an investment-grade rating. And when your business strategy depends on maintaining strong relationships with your suppliers and other constituencies, that sub-investment-grade rating could jeopardize your entire business model.

SOTER: I'm not sure companies need as much financial flexibility as they think they do. Let me give you an example. About a year ago, SPX Corporation, a New York Stock Exchange company (and also our client), announced a leveraged recapitalization. At the time, SPX had one issue of subordinated notes outstanding rated “Single-B.” The “corporate” credit rating assigned the company by Standard & Poor's was “BB-.” As part of the recapitalization, the company borrowed about $100 million to buy back 18% of its outstanding common stock in a Dutch auction. Within about 8 weeks of the announcement of the Dutch auction, the stock price had run from $43.50 right through the tender range of $48–$56, and settled at around $70. And, now that about 12 months have passed, it's currently trading at around $78.

Now, what about this issue of financial flexibility? First of all, SPX basically funded the entire transaction by completely eliminating the modest dividend that the company had been paying on its common stock. In fact, the cash flow savings from eliminating the dividend were actually larger than the after-tax cost of servicing the additional debt.

And, to take this issue of financial flexibility one step further, let me point out that SPX is now attempting to take over a company almost four times its size. Clearly, they're not going to be able to finance that transaction with bank lines. They're going to have to finance it in the marketplace. But, given their recent track record, it seems unlikely they will have any trouble financing the deal.

And that brings me back to the point that I was trying to make earlier—namely, that most managements don't like to be subjected to the scrutiny of the marketplace. What the story of SPX makes clear is that, provided you are willing to submit to the market test, you may not need much financial flexibility. As we were all taught in business school, you can raise capital on a project-by-project basis provided the market thinks the investments are promising. And the run-up in SPX's stock price—from about $75 to $78—since the announcement of the transaction suggests that the market has already accepted the deal.

But I can assure you that before SPX decided to undertake its leveraged recap 1 year ago, management was very concerned about financial risk and flexibility. And one of our biggest challenges in this case was convincing them of the limited significance, if not complete irrelevance, of bond ratings in the process of creating value.

HANS STOLL: Well, Dennis, I'm not as convinced as you seem to be that companies can always go to the capital markets when they have profitable opportunities. There are real problems arising from the so-called information asymmetry between management and investors—a possibility that could make financial flexibility quite valuable in some circumstances. If your company is fully levered, and your earnings have turned down for reasons beyond your control, you may be forced to pass up some profitable investments. Convincing the capital markets to provide funds in such cases could be very costly.

Alice, is that one major reason why companies want financial flexibility?

PETERSON: Well, that's part of the explanation. But it goes further than the here and now. Every company is different, and some of the middle market companies that Dennis deals with might very appropriately have the perspective that he has described. At Sears, we are very conscious of having been in business for 110 years, and we're planning to be around for at least another 110 years. And when you think about what you need to do to ensure that that happens, you come to appreciate that financial conservatism allows you to live through all kinds of economic cycles and competitive changes that could affect your company.

Much of our sense of the company comes from the knowledge of its accomplishments, from our sense of being part of a team that has done big things. We started Allstate from scratch in 1931, and we started the Discover Card from scratch in the late 1980s. The treasurer of a company must understand the capital resources (and their costs) needed to carry out the strategic plan and to support the business model. And that requires a degree of financing flexibility that is not available to firms with the highly leveraged balance sheets that Dennis has described.

MYERS: I'm with Alice on this one. I don't deny that there are situations in which the medicine of debt and restructuring is very apt and is very successful. But there are other situations where you don't want the company to go on a “diet,” or not at least on a “crash diet.” And, in such cases, you certainly don't want to let your thinking be driven by a second-order concern like taxes.

The first consideration should always be the investment opportunities, the business opportunities. The second consideration is the financial structure of the business. The third concern—and it's a distant third in my view—is the percentage mix of debt and equity on the balance sheet.

When you do get to the issue of the percentages on the balance sheet, taxes are only one of four or five things we know are potentially important. For this reason, I think it's sort of back-asswards to start off with taxes and say that's the primary concern, even though we can agree that it is one important element.

My second observation, Alice, if I were to put your speech in academic lingo…

PETERSON: By all means.

MYERS: …would go something like this: Economic theory is making a big mistake in treating all the employees of the corporation as temporary workers. As I suggested earlier, to succeed a corporation requires a coinvestment of financial capital from the outside and human capital that is built up inside the business. You need the human capital to make the business work in the long run. And it's not necessarily inefficient for people who have most of their human capital tied up in the business to be conservative in protecting it. Where human capital is fungible and people move all around—say, in the case of swaps traders—companies can afford to be much more aggressive in taking financial risks. And the employees won't care. But, in cases where there's lots of human capital really locked up in an organization—cases, for example, where somebody spends his or her whole career at a company—it's understandable that they would want to protect that investment.

STERN: Well, Stew, I'd like to make a suggestion to you. I think that what you're saying is almost certainly true— especially given the existing distribution between fixed and variable pay that exists in most public companies. If you take a look at people in middle management and below, their variable- pay component is typically a very modest amount. But, when you put them on an EVA incentive system— one where there are no caps on their awards and there is a bonus bank system to ensure that their payments are based on sustained improvements in performance—their behavior changes … almost overnight.

In some cases, they become in favor of a more aggressive capital structure because that reduces their cost of capital and increases their EVA. In addition to the case of SPX, which Dennis just mentioned, other companies such as Equifax and Briggs and Stratton have done leveraged recapitalizations after first going on an EVA program. I would also suggest that, after going on EVA, companies in so-called mature industries often begin acting as if their industries were not as mature as management seemed to think they were. That is, companies whose rates of return were well below their cost of capital for a long time suddenly became superior performers.

MYERS: That's great, Joel. But, once again, you are making my point. What you're saying is that by going in and changing financial structure—which includes the system of incentives inside the business—you can get the people who contribute human capital to the organization to take additional risks because they're given a reward.

STERN: Yes, that's right.

MYERS: My point, though, is that there's an investment in human capital that, from an efficiency point of view, you will want to protect much of the time. Now, you don't want to protect it when it's ceased to be productive—and that happens in many large bureaucratic organizations. But, by and large, it is economically efficient that when you ask people to make an investment of human capital in your firm, you do not then do things—like raising the leverage ratio too high—that would needlessly put that investment at risk.

STERN: Forgive me, Stew, but I still have a problem with this. Let's assume we do all of the things in the sequence that you propose, and that the issues of capital structure and taxes are the last things on the list. And let's assume that we get the 20 incentives right, and that we carry them down deep into the company.

Now, having done those things, we once again come to this issue about whether debt is not significantly less expensive than equity. And, as you suggest, Stew, in many cases we may come to the conclusion that the company cannot support much leverage, whether because the debt would endanger the strategic plan or put the human capital investment at undue risk. My argument is this: Even under these circumstances, I still suspect that there are a good many companies where lots of talented, energetic people would be more than willing to put their human capital at risk for the right payoffs.

After all, this is essentially what KKR does when they go into a company. They say to operating management, “If you're willing to take some additional risk, and subject yourself to very challenging performance standards, we can make it very rewarding for you.” As we now know from the research, the same operating managers have shown themselves capable of doubling their companies’ operating cash flow in a period of 3 years or less when you change the incentive system. And this is basically what we're trying to accomplish with EVA, but without the risk to human capital imposed by high leverage.

In a conference we held a number of weeks ago for our EVA clients, Michael Jensen and I got into an intellectual fist-fight on a favorite subject of his. Both Mike and my partner, Bennett Stewart, are fond of talking about the effectiveness of high debt ratios in discouraging managers from doing foolish things with shareholder resources. The basic idea is that debt exerts a discipline on management that can be valuable in companies with too much cash and few profitable investment opportunities.

As we have argued in a series of articles, our EVA-based incentive compensation is also designed to deal with this corporate “free cash flow” problem. And, because our system can be taken well down into the corporate structure—that is, into the individual business units and below—I would argue that EVA might even be more cost-effective than debt financing in discouraging the corporate waste of capital.

So, to the extent that we have succeeded in designing and implementing a measurement and reward system that motivates managers to make the most efficient use of capital—and I recognize that this is the critical assumption—then why would it be necessary to resort to debt to accomplish the same goal? That is, assuming debt creates a valuable discipline for management in certain cases, why wouldn't the “localized” incentives provided by an EVA plan be expected to do an even better job? After all, as you have pointed out, Stew, too much debt could kill all corporate investment, good as well as bad. An EVA system encourages managers to fight for just those projects that promise to earn their cost of capital.

SOTER: Let me take a crack at that one, Joel. At the risk of being asked to leave the partnership, let me tell you the following story. In 1992 we were helping Equifax, the nation's largest provider of consumer financial information, to adopt EVA incentives. At the same time, we were asked to advise the company on what turned out to be a rather aggressive leveraged recapitalization.

As the chief financial officer told me, “Look, anything can happen. There is no contractual obligation on the part of the board to continue to have EVA incentives here in this company. But the contractual obligations associated with debt are real, and the consequences of failing to meet them are far more weighty to the company and to the stockholders than canceling an EVA plan.”

As this statement suggests, there is an important difference between the contractual nature of the company having to service debt versus the obligations under EVA incentives. And the company's willingness to bind itself in this way sends a strong signal to investors. It demonstrates to the marketplace that agency problems are being addressed inside the company, and it can be very effective in eliminating at least the perception, if not the reality, of corporate reinvestment risk.

STERN: I have two responses, Dennis. First of all, I accept your resignation. The second is, if you take a look at recent announcements of public companies of just their intent to go on an EVA plan, the market response has been noticeably positive. In some cases, we have seen share values change by as much as 25% within a week's time.

Last week I took part in a conference at Stanford law school, and on the panel sitting next to me was George Roberts of KKR. I went through a rather lengthy discussion of the content, the design structure, and the outcomes of EVA plans. And when George was asked to comment on it, he said: “That sounds like a better idea than the one we've got. Why would you want to use leverage unless you really have to? The discipline of debt is a very blunt instrument to accomplish what you are saying you can do unit by unit throughout the organization.”

You see, we can even design the incentive system to encourage teamwork among the different business units while preserving rewards for individual performance. We could say to an operating head, “Seventy percent of the annual reward will be based on the performance of your unit, and 30% will be based on the performance of all the other units.” This way, if good ideas are developed in one part of the organization that could help other parts, they will quickly move from unit to unit.

So, what I am really suggesting, then, is that EVA can provide a solution to this human capital problem that Stew has mentioned as a reason for avoiding high leverage. We can take managers and employees with large investments in human capital and make them rationally less risk averse by making them partners in creating sustainable improvements in value. This way, managers and the existing shareholders can end up the big winner, and not somebody like KKR that comes in from the outside.

HANS STOLL: Up to this point, we have discussed capital structure policy as if it were designed primarily to reduce taxes and agency costs while also holding down expected costs of financial distress. But we have said very little about information costs. As Joel mentioned earlier, in 1984 Stew Myers presented a theory of capital structure called the “pecking order” theory, which relies heavily on information costs to explain corporate behavior. And I was wondering, Stew, if you could share with us your current thinking on the role of information costs in corporate financing decisions?

MYERS: The pecking order starts out with the prediction that companies will issue debt instead of equity most of the time because of information problems. If the information problems are important, then you're going to see the debt ratio vary across time according to companies’ needs for funds. Companies with a balance of payments deficit with respect to the outside world will be increasing their debt ratios while those that have a surplus will be paying down debt and reducing leverage.

I like to put it just that baldly, not because I think it's the complete answer, but because I want to get away from academics’ “nesting instinct.” The tendency of academics is to begin by observing that there are four or five things that could affect capital structure. Therefore, they say, let's find ten proxies for these four or five factors and run a regression. But, because each proxy has some connection to two or three of the underlying variables, you end up not being able to interpret the regression. Or you interpret it the way you like to interpret it, but not against an alternative theory.

Lakshmi Shyam-Sunder and I have a paper coming out shortly in the Journal of Financial Economics where we show that, for a sample of relatively mature and established public companies, the pecking order works great. We get R²s up around 0.80 in a time series. We also show that if it were true that companies were really trying to move to a target capital structure based on a tradeoff of taxes and cost of financial distress, you could reject the pecking order. Therefore, our test has power. At the same time, we show that if the companies were actually following the pecking order, it would look like they were seeking out some debt-equity target.

So if you want to start with one description of how large, established companies behave, why not start with the one that has a very high R² and for which we can show statistical power?

SOTER: Is the pecking order theory consistent with value-maximizing behavior?

MYERS: Yes.

SOTER: But as I understand it, one of the premises of the pecking order theory is that dividends are sticky. And, for that reason alone, I have trouble seeing how that policy can be consistent with value-maximizing behavior.

MYERS: Are we talking about dividend policy now?

SOTER: I don't see how you can separate the two. In my experience, they are interdependent decisions— or at least we encourage management to view them that way.

MYERS: Why do dividends matter aside from taxes?

SOTER: If they don't matter aside from taxes, then my first thought would be: Why don't we cut back on dividends before we even seek or consider seeking external financing? This is what we did in the case of SPX that I just mentioned.

MYERS: I don't have a complete answer to that, and I think that's one of the biggest unsolved problems in academic finance. We don't have a good theory of dividend policy. We have no theory of dividend policy that goes back to any kind of fundamentals. And so I admit the pecking order takes the stickiness of dividends as a fact. It doesn't try to explain it. I don't think anybody can explain it.

SOTER: Let me give you one other example of how a dividend cut can be used to facilitate an increase in leverage. A year ago, we advised IPALCO Enterprises, the parent company of Indianapolis Power and Light, in becoming the first utility ever to undertake a leveraged recapitalization. IPALCO started off with a “AA” bond rating and a 42% debt-to-capital ratio. The company borrowed more than $400 million and used the proceeds to buy back its common stock. In the process, the company's debt-to-capital ratio went from 42% to 69%. But the increase in debt was financed in effect with an almost 33% cut in the dividend, from $1.48 per share to $1 per share. In fact, the annual savings from the dividend cut of about $41 million exceeded the increase in the after-tax interest expense by about $25 million. That is, IPALCO's after-tax cash flows actually increased by $25 million even as the leverage ratio increased from 42% to 69%. And both of the rating agencies the day after the announcement reaffirmed the company's “AA-”/ “AA” bond ratings, which further suggests that leverage ratios, at least in traditional book accounting terms, just don't matter.

IPALCO's operating cash flows have also increased since that time, and the stock market seems to have liked what's happened. In 1997, the company's stock provided a total shareholder return of 58.5%, putting it among the top three electric utilities last year.

McCONNELL: I want to just stop and take stock for a moment, and to make sure that, if there is a debate here, we all understand what the debate is about. If I understand Stew's observation, it is that capital structure may matter but it doesn't seem to matter very much, at least from a manager's perspective. Managers don't spend a lot of time worrying about it. Is that a reasonable characterization?

MYERS: That's right, at least for large, established companies within a wide middle range of debt ratios.

McCONNELL: What Dennis is arguing, if I understand it, is that capital structure definitely matters, and he has given us several case illustrations. One problem with the use of cases, of course, is that you hear only about the ones that work—which is not to say, Dennis, that all of yours didn't work; I'm sure they did.

Ron Masulis has done a classic study of the stock market's response to exchange offers. And that study shows very clearly—and for a large sample of companies, not just one or two interesting cases—that when companies announce that they are issuing new debt to retire stock, their stock price increases. And the study also shows the converse: when stock is issued to retire debt, stock prices decline. Now, the question is: Why does that happen? How should we interpret the market's reaction?

Some people have argued that there is a signaling effect at work here. That is, most companies tend to do leverage-increasing recaps only when they expect earnings—or, more precisely, operating cash flows—to increase in the future. To the extent investors understand this, they raise stock prices in anticipation of higher future operating earnings. This may in fact explain the market's reaction to the leveraged recaps that Dennis described.

And the opposite story is likely to explain the market's negative response to leverage-reducing recaps. If you look closely at Ron's data, what you find is that 90% of the companies that do leverage-reducing recaps are actually in some degree of financial distress. So, what we may be picking up from the announcement of an equity increasing swap is a signal that we're in real trouble, guys.

My question, Dennis, is this: Are these leveraged recaps valuable in and of themselves—say, for their tax savings and for some beneficial incentive effect? Or are they simply functioning as signals of management's confidence in future earnings?

SOTER: You are certainly right to warn that I could be dealing with a biased sample, and I realize that anecdotal evidence does not constitute proof. But let me respond to your question by telling you a little more about IPALCO.

McCONNELL: Go ahead. I've been at Purdue for 20 years now, and it's close to my heart.

SOTER: Since it's a utility, Indianapolis Power and Light probably goes well beyond what most unregulated companies provide in terms of disclosure. In fact, for all practical purposes, they share with the investment community a 5-year plan. Utility analysts are famous for forecasting earnings per share within one penny each quarter because management is spoon-feeding them all the information. So, I don't think there was much room for signaling associated with their new financial strategy.

On the other hand, I think you can argue that there was a lot signaling in the cases of SPX and Briggs and Stratton. In the case of Briggs and Stratton, the stock price went from $42.75 to $48 during the week of the announcement. And, as I mentioned earlier, SPX's stock price started at $43.50, ran right through the price range for the Dutch auction (of $48 to $56) and settled at close to $70.

But let me also make the following observation: If signaling is the explanation, there are almost certain to be a number of other, real effects behind the signal. For example, both SPX and Briggs and Stratton had been EVA companies for some time before making announcing their recaps. And, since investors had already achieved a degree of confidence in these two management teams, this signal of future improvements coming on top of recent past achievements could be especially convincing.

And, to show you what I mean by real improvements, let me now go back to the case of IPALCO. About 7 months after IPALCO completed its recap, I had lunch with the chief financial officer, John Brehm. In describing some of the events that had happened internally since the recap, he said, “It's been extraordinary. Through 8 months of 1997, we have been right on budget with respect to reported earnings per share, but we are $40 million ahead on cash flow since the recap.” He said that the recap had really focused people on generating cash flow. They are running the business differently, even though it has not yet been reflected in reported earnings. (And I must confess that even I was a little surprised; I didn't think anything would have focused the attention of utility managers.)

So, I think there are several things at work here. Yes, there is some signaling. But you have to ask: What's being signaled—or why is management more confident about future operating cash flow? Well, a big part of the answer is that managers now have stronger incentives to produce operating cash flow, and debt helps increase cash flow by sheltering corporate taxes. And, on top of all this, the substitution of stock buybacks for cash dividends has the effect of increasing after-tax rates of return to stockholders by distributing cash in a more tax-efficient way.

MYERS: There is a signal in the utility business, by the way. If I were a utility investor, my worry would be that the utility would take all this cash flow that's coming into its pockets and burn it somewhere. And the leveraged recap acts as a signal, in your case, that they are not going to do that.

PETERSON: I agree that, just as there's nothing like the prospect of the guillotine to concentrate the mind, leverage can have wonderful effects in certain circumstances. But I also believe that we can create incentives inside companies that produce the LBO mindset and the behavior that goes with it without piling on leverage. In other words, I think we have to think about how we get the behavior we want, and I think there are a lot of ways to do it.

The foolproof way, as Joel has pointed out, is to lever up and require managers to own lots of stock. There is nothing like having that debt service staring them in the face every day. So I agree with that point.

STERN: Alice, what is the feeling about dividend policy at Sears? Could Sears cut its dividend by a third, or even to zero, and borrow a lot more and thereby increase its value?

PETERSON: We haven't increased our dividend for more than 10 years. In fact, we have effectively cut it by virtue of the spinoffs that we've done. When we spun off Dean Witter Discover, we reduced our total dividend payments; and when we spun off Allstate, we reduced it again.

But that, of course, is not the same thing as cutting it to zero. And that is an action with very serious consequences. Sears is one of the most actively traded stocks in the United States. And, when you have a stock that is as actively traded and as broadly held as ours, if you decide to eliminate your dividend you must be prepared to deal with a very dramatic shift in your stockholder base. I'm not saying that it couldn't be done. But it would take a great deal of preparation, and it would have to be communicated very, very carefully to the investment community. And it's not something that we plan to do any time in the near future.

In other words, positioning your company for the stock market is an extremely critical part of the whole circle of activities and decisions that turn on how much flexibility you want as a company. As in the case of leverage, you need to set your dividend policy in such a way that it supports your business model, your ability to carry out your strategic plan.

SOTER: I'm not as convinced that dividend policy needs to be established with a concern for the needs of the current stockholder base. In 1994, we advised the board of FPL Group, the parent company of Florida Power & Light, in its decision to commit what in the minds of many was financial hara-kiri. FPL was the first profitable utility, to my knowledge, to voluntarily cut its dividend. The dividend was cut nearly one-third, from a quarterly payout of 62 cents to 42 cents per share; and in its place the board authorized management to repurchase up to 10 million shares of common stock in the open market. If ever there was a test of the so-called “clientele effect,” this was it. I recall a director at the board meeting that preceded the final decision to make the cut asking the question, “What's the likely impact on our stock price?” I responded— conservatively I hoped—that the market price would drop up to 25%, and would not fully recover for as long as 2 or 3 months.

Sure enough, the stock dropped 15% on the day of the announcement. But within 3 weeks the price had fully recovered. What's more, the dividend cut itself caused 15 brokerage houses to put a “buy” on the stock. And the stock attracted lots of new institutional investors, even as it lost many others. During the 12 months following the dividend action, FPL's stockholders realized a total return of 23.8%, more than double the S&P Electric Index. So, there was a significant change in the investor clientele, in the composition of the stockholder base.

And this also tends to happens with leveraged recapitalizations. When we advised Equifax in 1992 in a very aggressive leveraged recapitalization, there was also a significant change in the investor make-up of the company. These two experiences lead me to believe that one stockholder clientele is indeed as good as another.

MYERS: Your case is interesting, Dennis, but I'm not sure it has much to say 24 to a company like Sears. Suppose Sears cuts its dividend to zero, but without taking a leveraged recap. All it's going to do is to tell shareholders, “Don't worry, we'll use the money we would have used for the dividend to buy back shares over time?”

SOTER: That's exactly what Florida Power & Light did. They did not do a leveraged recapitalization, but they did announce that they would be using part of the cash savings to buy back their shares in the open market. So, it's a nice, almost laboratory-controlled, example of where we cut cash dividends and substituted a policy of open market repurchases.

PETERSON: Our payout ratio is in the 20%–25% range, which is a little bit lower than that of our peer group. Another consideration—although not necessarily the most important factor— is that our shareholders are often our customers. These so-called “retail” investors tend to want dividends.

MYERS: I tend to agree with Alice's position. And I would think even the most zealous market-efficiency types should not have trouble with it. The “perfect markets” view of finance would say that, aside from temporary signaling effects, it doesn't really matter very much whether you pay dividends or buy back shares. And so, if somebody says, “Why pay dividends?,” the standard answer is, “Why not? There are clienteles out there that want dividends, and consumers are sovereign.”

So, if somebody was going to pay dividends in this world, why wouldn't Sears be likely to satisfy that demand? I don't see anything wrong with that argument. The only possible objection I can see would be taxes again. But I don't think anybody has proved that high dividend yields lead to a higher cost of capital because of taxes.

STERN: Let me ask you a question, then, Stew. If what you're saying is true, then why don't some companies have two classes of shares, one that would pay a high dividend and one would pay a low dividend? This way, the shareholders could buy whichever mix of those shares they wanted so as to write their own dividend policy. If dividends were so terribly important that there was an optimal dividend policy for investors, then you would expect that there would be more than one class of share available to investors so that they could manufacture their own dividend policy.

MYERS: You didn't hear what I said. I said that dividend policy is not important, aside from the short-run signals that are given. It doesn't matter whether cash is paid out via dividends or share repurchases. (The total amount of cash paid out does matter, of course.)

But there are people out there who, for whatever reasons, want some dividends. Somebody has got to give it to them. It doesn't cost Sears very much to give it to them. They are natural provider of that service. Why not? If somebody else wants to repurchase shares, good for them. I just don't care.

SOTER: What if I were to suggest that here, too, we can have our cake and eat it, too. Before joining Stern Stewart, as Joel mentioned, I worked for Citizens Utilities. The founder of Citizens Utilities was Richard Rosenthal, and some of you may be aware of this case because, back in the ’70s, John Long at the University of Rochester wrote a paper about the company's highly unusual dividend policy.

For many years, Citizens had two classes of common stock, one that paid a cash dividend, and one that paid a common stock dividend of an equivalent amount. Richard had gotten this grandfathered originally in 1956 and, through successive tax acts, that waiver had been extended. But, in 1990 as the Bush tax act was coming up, we saw that we were going to lose the exemption. So we had two options: We could eliminate the stock dividend and change it to a cash dividend, or we could convert the cash dividend class to a stock dividend, and so eliminate cash dividends altogether.

We chose the second course. We obviously weren't going to start paying cash dividends on the stock dividend class. To soften the blow for the stockholders who wanted cash, we created the “stock dividend sale plan.” We said, in effect, “We're going to give all stockholders stock dividends, but those stockholders who want to receive cash income can make an election that effectively enables them to convert those dividends into cash.” You see, the company couldn't just buy back their shares because the IRS would have viewed that as a sham, and the cash would have been taxed as ordinary income. So, instead we simply made arrangements with an investment bank and our transfer agent to aggregate those shares that stockholders elected to cash in to maintain their income stream—and those shares were sold, with no commission costs (except the bid-ask spreads), to provide stockholders with the same cash income. Our shareholders were taxed on that income at a capital gains tax rate, and only on the difference between the sales price and their basis.

So, we achieved a significant reduction in taxes for the stockholders who elected to continue to receive income.

MYERS: It's good financial engineering. But, again, I think it's a second order effect, just as Miller & Modigliani taught us.

STERN: Well, I think that's a good note on which to bring this to a close, and I want to thank you all very much for your participation. We expected this to be an exciting program, and I think we got our fair rate of return.