Search GDAE

Subscribe for
E-mail Updates

Recent Publications

GDAE in the News

Upcoming Events

Publicaciones en Espanol

Jobs and Resources

• Aggregate rates of investment in property, plant and equipment, civilian R&D, and intangible assets such as corporate training and related forms of corporate human resource development is lower in the United States than in Japan and Germany (see Farb 1992).
• Anecdotal evidence suggests that U.S. firms invest at a lower rate than their German and Japanese counterparts in nontraditional forms such as human resource development, relationships with suppliers, and start-up losses to enter foreign markets. Kochan and Osterman (1992) and Blinder and Krueger (1992), for example, both document a significantly lower rate of investment by U.S. firms in human resources.|4 The magnitude of the difference appears greater in this area than in other investment areas.
• The R&D portfolios of U.S. firms include a smaller share of long-term projects than those of Japanese and European firms. A survey of CEOs in the United States, Japan, and Europe by Poterba and Summers (1992) found that 21 percent of the projects of U.S. firms were considered long term compared with 47 percent in Japan and 61 percent in Europe.
• The average holding period of stocks has declined from over seven years in 1960 to about two years today (Duttweiler 1991). This decline implies a dramatic shift in the frequency with which investors buy and sell corporate equities. It is perhaps the most telling evidence of shortening investor horizons.

Although these findings present a broadly consistent picture of lagging U.S. investment, there are some puzzling and important complexities that seem to defy the overall pattern. These puzzles derail many simple explanations for why the United States invests less or has a shorter time horizon. For example, the U.S. investment (and competitiveness) problem varies by industry and even by company: the United States is highly competitive and invests aggressively in industries such as chemicals, pharmaceuticals, software, and telecommunications network equipment. The United States does well in high-risk, long time horizon start-ups and invests heavily in emerging industries. Finally, many U.S. firms seem to have overinvested in unrelated acquisitions. Any useful understanding of the broad problems we are addressing must also take account of these apparent counterexamples.

The Determinants of Investment

The determinants of investment can be grouped into three broad categories: the macroeconomic environment in which all companies operate; the allocation mechanisms by which capital moves from its holders to investment projects (it is on this second category that our research focuses); and the conditions surrounding specific investment opportunities themselves. (These latter include characteristics of the particular projects, companies, industries, and the geographic locations where investments take place.)

Investment capital is allocated via two distinct but related markets: the external capital market through which holders of equity and debt provide capital to particular companies and the internal capital market in which companies allocate the internally and externally generated funds at their disposal to particular investment programs. There are strong parallels in how these dual markets function. The equity holder's task of selecting and monitoring companies in which to invest bears a strong resemblance to the corporation's task of selecting and monitoring investment programs. Both markets also share attributes that will prove crucial to investment behavior: highly incomplete information that is costly to gather; the existence of principal-agent relationships (in some cases, multitiered ones) that create agency costs; important spillovers or externalities in investment; and the potential to influence the results of a given investment through intervention.

The Nature of the External Capital Market

The external capital market includes a number of entities. The first is the holders of equity capital, whom we will term owners. The second is the agents who, in some instances, make investment choices on the owners' behalf. In the United States, investment decisions involving a large fraction of equity capital are made by institutional agents, whether they are pension funds, mutual funds, bank trust departments, or other money managers. Because owners have incomplete information about the performance of agents, they must use proxies in monitoring and evaluation. Agents can be expected to manage investments in a way that is aligned with how they are measured and compensated.

A third important entity in the external capital market is lenders of debt capital. The fourth entity is, of course, the corporation in which equity is invested or to which a loan is made. The final important entity is the board of directors, which is the formal link between the owners and the corporation.

The way in which the national system for capital allocation deals with the problem of imperfect information is crucial to understanding investment behavior. How owners/agents cope with imperfect information in dealing with each other and in valuing companies will influence the particular corporate attributes that influence stock prices. Management's perception of the ways in which the external market values their companies affects internal investment choices. The way in which managers cope with imperfect information internally will influence the particular types of investment projects selected. Finally, the approaches used by owners/agents to deal with incomplete information will influence those used by management.

Four characteristics of the external capital market are of principal importance for investment behavior. The first is the pattern of share ownership and agency relationships, which refers to the nature of the owners, the extent of their representation by agents, and the size of the stakes held in companies. The second is the goals of owners and agents, which influence their desired investment outcomes. The ability to jointly hold debt and equity is one important influence on goals, as is the existence of a principal-agent relationship. The third salient characteristic of the external market is the approach and information used by owners/agents to monitor (measure) and value companies. The final important characteristic is the ways in which owners/agents can influence management behavior in the companies whose shares they own.

These separate characteristics of the external capital market in an economy are interrelated and will tend over time to become internally consistent. If the dominant investors are principals who hold large stakes and are permanent, relationship-driven owners, the incentive will be present to invest in and accumulate fundamental knowledge on long-term corporate prospects. Owners will demand and achieve more influence on management behavior. Their orientation will be to work with management to improve performance and sell out only as a last resort. Conversely, if the pattern of share ownership/agency relationship is dominated by transaction-driven agents holding small company stakes who are measured on near-term appreciation and are dependent on public information, the incentive will be to employ information-lean index funds or simple value proxies in monitoring and valuation rather than to invest in costly fundamental research. Share holdings will tend to be more transient if agents can exert little influence on management.

The U.S. External Capital Market

In the United States, publicly traded companies increasingly have a transient ownership base comprised of institutional investors such as pension funds, mutual funds, or other money managers that act as agents for individual investors (see Edwards and Eisenbeis 1992). Such agents account for at least 55 percent of the public market and, combined with member firms trading on their own account, over two-thirds of all trading.

Due to legal constraints on concentrated ownership, fiduciary requirements that encourage extensive diversification, and a strong desire for liquidity, U.S. institutional agents hold portfolios involving small stakes in many, if not hundreds, of companies. For example, one of the largest U.S. pension funds, the California Public Employees Retirement System (CalPERS) reportedly held stock in over 2,000 U.S. companies in 1990, and its largest individual stake was 0.71 percent (see White 1991; Sailer 1991). Due to restrictions on bank ownership of corporate equity, U.S. banks do not normally hold debt and equity in the same company. Neither do U.S. insurance companies, although they are not restricted from doing so.

U.S. institutional agents such as pension funds and mutual funds, and the managers of such funds, are typically measured on quarterly or annual appreciation compared to stock indexes, and they thus seek near-term appreciation of their shares. Quarterly performance evaluation encourages the documented practices of "window dressing" (selling poorly performing stocks at the end of the year) and "lock-in" (well-performing portfolios are sold and the investment managers buy S&P 500 to lock in gains relative to other funds), although these practices have limited effect on returns to shareholders. Managers are readily changed if performance is unsatisfactory. Pension funds, whose capital gains are not taxed, make buy/sell choices without considering the trade-off of having to pay taxes on appreciated stocks sold, which would encourage longer-term holding. Mutual funds must distribute income and capital gains annually so that investors pay taxes each year. This only serves to heighten the attention of funds to quarterly or annual realized performance and hence to realizing gains. As a result of all these circumstances, mutual funds and actively managed pension funds (which represent 80 percent of pension assets) hold shares, on average, for just 1.9 years.|5

The situation just described -- in which institutional investors have fragmented stakes in numerous companies, short expected holding periods, and lack of access to "inside" information through disclosure or board membership -- results in a strong tendency to base buy/sell choices on relatively limited information, based on those measurable company or industry attributes that affect near-term stock price movements.

Current earnings are an example of a proxy for corporate value. Their use is problematical, first, because accounting earnings do not accurately measure true earnings. Many investments must be expensed, and as a result current accounting earnings understate true earnings in companies where high rates of intangible investment are needed. Second, even true current earnings do not accurately measure corporate value because they fail to measure the firm's competitive position and its ability to sustain or improve that position and hence its future earning power. Maximizing a series of short-term returns is not equivalent to maximizing long-term returns because of the role of multiperiod investments. Investments needed to build skills, capabilities, and market positions, or to defend existing positions at the expense of current earnings are penalized by the use of current earnings as a value proxy.

Despite their large aggregate holdings, U.S. institutional agents have virtually no real influence on management behavior. For various legal and regulatory reasons that will be discussed later, agents rarely hold seats on the board of directors and have little clout with management. Index funds, which might be seen as long-term investors, cannot play this role effectively either. With their investment philosophy, extreme fragmentation of ownership, and lack of incentive to invest in information, index funds have little realistic prospects of credibly monitoring and influencing management behavior.

Institutional agents are left, then, with the proxy system as the only direct means of imposing their views on management. The proxy system, while cumbersome and ineffective, has been employed with growing frequency in recent years, with the number of shareholder proposals increasing from 33 in 1987 to 153 in 1990. The great majority (91 percent) of shareholder proposals regarding economic matters, however, have little to do with strategy or investment behavior but relate to corporate policies affecting the ease of control changes. Here, shareholder proposals invariably aim to make takeovers easier. Given the goals of institutional agents and their approach to monitoring and valuation, this is not surprising. Yet many institutions do not even bother to vote their shares, and there is growing debate about the responsibility of institutions in voting proxies as agents for investors.

The only real remedy available to U.S. institutional agents who are dissatisfied with management decisions is to sell their shares. The stock prices of out-of-favor companies are bid down by investors seeking to limit their losses who are unable to wait (because of their need for rapid appreciation) for a rebound due to earnings improvement or for a possible control change. Successive stock price declines may ultimately create the potential for acquisition or takeover to unlock the unrealized value. Arbitrageurs, a largely different group of investors from institutional agents, provide a countervailing force that holds up share prices of companies for which takeover seems likely.

Control changes and voluntary restructurings to forestall them, then, represent the only real discipline on management behavior in the U.S. system, given the limited power of owners and the limited impact of the proxy system. Unfortunately, this form of discipline occurs only after protracted decline in corporate performance. It also involves high transactions costs and carries with it some potential negatives for long-term corporate performance. While takeovers and buyouts can reduce static inefficiency through forced restructuring, they replace institutional agents with another financially oriented owner, the LBO firm (ironically, the funding for takeovers often comes from the very institutional agents who were the previous shareholders). While the new owner may have a longer horizon than the previous ones, most LBO firms are prone to being transaction-driven and seek to realize gains from selling out or taking the company public. In addition, the leverage required to complete the takeover forces asset sales that may in some cases diminish competitive capabilities and instill a financial stringency that can make it difficult to fund investments that do not generate cash quickly, particularly unexpected ones.

The Internal Capital Market

The internal capital market is the system by which corporations allocate the capital available from both internal and external sources among competing investment projects within and across business units. The system of internal capital allocation in an economy mirrors and is significantly influenced by the external capital allocation system. The external market sends signals that affect how companies are organized and managed. However, there are also independent influences on the internal capital allocation process that arise from laws, regulations, and prevailing management practices.

The most important influences on the internal capital market parallel those that shape the external market: the particular goals corporations set, the organizational principles that govern the relationship between senior management and units, the particular information and methods used to value and monitor internal investment options, and the nature of interventions by senior management into investment projects. The issue that will be emphasized here is that of corporate goals. (For fuller exposition of the other issues, see Porter 1993.)

Corporate goals are influenced from a number of directions. One is the legal framework in a nation that defines corporate purpose, as well as the extent to which particular corporate goals are codified in law as the duties and responsibilities of directors and managers. Particularly significant to corporate goal setting is the extent to which shareholders' goals are given explicit legal primacy and how such laws are interpreted by participants in corporate governance and by the courts. The way such laws are interpreted and the risk of lawsuits are as important as the laws themselves because this influences the sensitivities of managers and directors and the tests applied in judging their behavior.

Also significant in corporate goal setting is the board of directors. The composition of the board defines the interests that board members represent and the knowledge they have available. A board consisting of major owners, for example, is prone to have a different orientation than one consisting solely of management. Also important in understanding the influence of the board on corporate goals is the board's power, in practice, to influence management choices. Power is a function of the information available to directors, their independence from management, and the extent to which directors speak for significant owners.

A final major influence on corporate goal setting is the way in which senior management is rewarded, notably its basis for compensation and promotion. A management that is compensated heavily based on current period accounting earnings, for example, will set different goals than one compensated based on market share. Similarly, a management that expects to hold its position for a decade sets different goals than one hoping to be promoted quickly or moved to a position in another business unit or company.

American corporate goals are centered on earning high returns on investment or maximizing "shareholder value," measured by the highly imperfect indicator of current stock price. While the prevailing legal interpretation allows for long-term shareholder value to outweigh short-term stock price in evaluating corporate decisions, the burden of proof has seemed clearly to rest in the opposite direction. Managements (and directors) who sacrifice short-term stock price for long-term shareholder value still run the risk of lawsuits, while companies that maximize short-term stock price at the expense of long-term shareholder value normally remain unscathed.

In the United States, the board of directors holds ultimate responsibility for corporate performance but exerts relatively limited influence on corporate goals or management actions. Nominees are selected by the current board. The CEO is often the source of ideas for new candidates and often exerts a strong influence over the election process as well as the ongoing operation of the board. This is partly because most CEOs are also board chairmen (in 83 percent of corporations surveyed by Lorsch and MacIver 1991). Recent growth in the size of boards has also weakened the power of directors by making discussion and decision making in formal meetings extremely difficult (see Johnson 1990).

The composition of the board further enhances the relative power of the CEO. Boards have come to be dominated by outside directors who exert limited influence on corporate goals, while board representation by major owners, bankers, customers, and suppliers has diminished. An estimated 74 percent of the directors of Fortune 1000 companies are now outsiders with no direct ties to the corporation, and 70 to 80 percent are full-time CEOs of their own companies (Eaton 1990; and Lorsch and MacIver 1991). Because employees, customers, and suppliers are typically viewed as having interests that conflict with the corporation they are rarely represented on boards and thus have limited influence on corporate goals.

Employee pension funds, which could serve to align the interests of employees and owners or represent employee views to management, are limited to holding 10 percent of their assets in the sponsoring firm. Pension funds are also essentially prohibited from holding seats on the board of directors. It is ironic that employee pension funds have come to embody the pressures of the external market even though they could provide a valuable role in balancing stakeholder interests.

The move to outside directors arose out of calls for greater board objectivity. Yet lack of ties by the director to the corporation limits the ability of directors to absorb the vast amounts of information required to understand a firm's internal operations. Moreover, directors typically hold rather limited ownership stakes. While the median aggregate holdings of the board account for an estimated 3.6 percent of equity, most directors have no shares at all or only nominal holdings (Lorsch and MacIver 1991; Morck, Shleifer, and Vishny 1988). In practice, then, neither owners, lenders, directors, employees, customers, nor suppliers exert direct influence on corporate goals.

Thus, the dominant influence on corporate goals is management, which interprets signals from the external capital market (perhaps incorrectly) and which is often subject to limited direct influence by either boards or owners. The goals set by U.S. managements are typically framed in terms of ROI (return on investment) or enhancing stock price. Managers care about stock prices because, among other things, they affect the corporation's ability to raise new equity and its vulnerability to control changes, which diminish their power or eliminate their position.

As we have seen, U.S. managements are oriented toward reporting high rates of return and responding to investor signals about how to maintain high stock price. Table 1.1, which reports the comparative ranking of goals in a sample of U.S. and Japanese companies, highlights the primacy of current rate of return and stock price among U.S. managements. In Japanese companies, earnings are significant because they are necessary to fund other priorities, but stock prices are essentially irrelevant.The pressures imposed by corporate goals have been exacerbated by a historical trend toward decentralized management structures. These began to proliferate as early as the 1950s, as wartime planning and budgeting techniques were adopted by corporate management and U.S. companies first reached huge scale and complexity. These developments began distancing senior management from the details of the business.

TABLE 1
Comparisons of Japanese and U.S. Corporate Objectives:
Mean Importance Rating

Source: Data from Kagawa, Nonaka, Sakakibara, and Okumura 1981.
Note: 3 = most important, 0 = least important

An explosion of unrelated or loosely related diversification, dating from the 1960s, both reflected and reinforced these tendencies.|6

Over time, information flow and top management involvement in business unit management have fallen, and extreme forms of decentralization involving little cross-business unit interchange have spread. Corporate leadership positions have come increasingly to be filled by nontechnical executives, and tenure in senior management positions seems to have decreased. With these changes, the view has developed that general managers can run any business, regardless of product line or technology. These organizational changes have been supported by, and have contributed to, an increased reliance on financial budgeting and other management systems that use quantitative techniques to evaluate the performance of business units and to make investment decisions.

A Comparative Perspective on the U.S. Problem

The cumulative and combined effects of the system that has been described here are especially visible in the comparison between U.S. investment behavior and the behavior of our Japanese and German counterparts. We may summarize the contrasts in the following ways.

• The U.S. system is less supportive of investment overall because of its greater sensitivity to current returns for most established companies, combined with corporate goals that stress current stock price over long-term corporate position.
• The U.S. system favors those forms of investment for which returns are most readily measurable, due to the importance of financial returns and the limited information available to investors and managers.
• The U.S. system favors discrete, stand-alone investments that generate leaps in position over ongoing investments required to build capabilities, those whose payoffs depend on complementary investments in other forms, or those that create an option value for the future.
• The U.S. system heavily favors acquisitions, which involve assets that can be easily valued, over internal development projects that are more difficult to value and constitute a drag on current earnings.
• The Japanese and German systems encourage aggressive investment in established businesses to upgrade capabilities and productivity. They also encourage investment in intangibles and internal diversification in order to redeploy personnel and secure the future of the enterprise. This comes at the cost, however, of a tendency to overinvest in capacity, to proliferate products, and to maintain unprofitable businesses indefinitely.|7 Especially in Japan, managers have discretion to make poor decisions as long as results are tolerable.
  Managers also have less incentive for strong individual performance in some respects, and these systems are prone to overemployment and greater difficulty in weeding out poorly performing employees. There is also an inability to rapidly enter emerging fields, especially via start-ups. However, unlike the U.S. firms in mature industries that invest their excessive cash in the wrong areas, Japanese firms benefit from the forces in the external and internal markets that tend to direct excess capital into the areas more aligned with long-term value. The benefits from overinvestment are thus higher.

If the U.S. system comes closer to optimizing short-term private returns, the Japanese and German systems appear to come closer to optimizing long-term private and social returns. They appear better able to address the investment spillovers within the firm. Goals that stress market position and corporate perpetuity and management processes involving greater top management information and cross-unit interchange seem better able to encourage complementary, shared, and option value-creating investments. Their greater focus on long-term corporate position, and an ownership structure and governance process that incorporates the interests of employees, suppliers, customers, and the local community allow the Japanese and German economies to better capture the social benefits that can be created by private investment.

The U.S. system for allocating investment capital has potent disadvantages, yet the Japanese and German systems are not ideal in every respect. There are important trade-offs among national systems. The U.S. system is good at reallocating capital among sectors, funding emerging fields, shifting resources out of "unprofitable" industries, and achieving high private returns each period, as measured by the United States' higher corporate returns on investment (Lawrence 1992).

This responsiveness and flexibility is achieved at the price of failing to invest enough to secure competitive positions in existing businesses and investing in the wrong forms. The U.S. system discourages investment in many companies and industries, particularly in intangible assets and in the linked and complementary investments that are needed to sustain and upgrade competitive capability. It is skewed toward acquisitions as opposed to building businesses internally, which leads to highly consolidated industries in which competitive rivalry is threatened. It also fosters overinvestment in profitable, mature sectors with few attractive investment opportunities.

2. The Case for Reform
Regulatory Influences on the U.S. System

To a considerable degree, the U.S. investment problem is our own creation. Many of the changes that have occurred and the trade-offs now present in the U.S. system have developed out of the regulatory regime established in the 1930s to deal with the perceived abuses occurring in financial markets at that time. In the intervening years, a long series of piecemeal regulatory choices have been made that have resulted in the U.S. investment system in its current form.

The great majority of the regulatory choices underlying the U.S. system were enacted to address goals that were often laudable but different from corporate investment behavior. The first principle, dating back to the 1930s or even before, was to avoid the concentration of economic power. The alleged abuses of large, dominant owners of companies were seen as an important cause of the Depression. U.S. regulators sought to limit the power of financial institutions, in particular, to affect stock prices or to manipulate corporate behavior.

A second and related principle was to separate the investment and control functions. The view emerged that there was a conflict of interest between the role of an investment manager and the role of a controlling owner. Controlling owners could take actions that could hurt individual shareholders such as dumping worthless shares into their managed investment portfolios. Similarly, exercising influence as an owner through such vehicles as a role on the board of directors was seen as compromising the objectivity of an investment manager.

A third guiding principle of U.S. regulation has been equal treatment of all investors, big and small, by providing for equal disclosure of material information and protection against abuses by insiders. Closely related to equal treatment was a fourth principle: to protect small investors, pension holders, insurance policyholders, and bank depositors.

The principles guiding U.S. regulation address some legitimate and commendable purposes and have achieved the goal of keeping abuses to a bare minimum. The result is public markets that are superior in some important respects to those in Japan and Germany. However, the cumulative pattern of regulation has had unintended consequences for investment behavior in the U.S. economy. A number of laws and regulations have directly or indirectly ensured that U.S. institutional agents hold small stakes in companies and that their holdings are widely diversified. Mutual funds, for example, are discouraged by tax incentives for diversification in the Internal Revenue Code and by reporting requirements under the Investment Company Act, from owning more than 10 percent of the stock of any firm. The fiduciary responsibility of pension fund and trust fund managers requires "prudent" diversification. In practice, this means
that managers hold small stakes in dozens if not hundreds of companies. State law governing life insurance companies also requires significant diversification.

By stimulating excessive diversification and the holding of many small stakes in companies, the U.S. system has encouraged frequent trading and heightened the influence of accounting earnings on buy/sell choices. This, in turn, made timely disclosure of event information crucial and insider trading a major concern, compared to a system in which share ownership was more stable.

Disclosure rules prohibit significant owners from becoming knowledgeable "insiders," while limits on board membership bar them from direct access to in-depth company information. While there is a great deal of information that leads almost instantaneously to significant swings in share prices, the information is of the wrong type to match capital with those companies with the best long-term prospects. The ultimate absurdity is that U.S. institutions are driven to index funds, which are information free, or to simple value proxy investing involving little or no fundamental information or concern about long-term company prospects as the best available investment alternatives.

In most cases, the abuses that most concerned regulators occurred in situations where there was unequal power and goals that were not aligned. The more powerful interest then took advantage of the other. Rather than better align the goals, however, regulators sought to equalize the power in the system by simply eliminating powerful owners or severely restricting their activities. Ironically, the result was that corporate managers became relatively free from direct influence or oversight by major owners, while they were still reliant on capital from owners whose goals increasingly diverged from those of the corporation.

The record shows a near total failure by legislators from the 1930s to the 1980s to consider the effects of regulation on corporate performance. Moreover, each successive area of regulation often led to more regulation. Due to the tendency toward mutual consistency, regulations in one area can also shape other parts of the system. Similarly, inconsistent regulations of different parts of the system can produce conflicts and frustration. A common outcome of such conflicts is further regulation.

Moving to Reform the System

Improving the U.S. system for capital allocation will require complementary changes in public policy, in the behavior of institutional investors, and in the practices of management. To be most effective, an array of changes must be implemented simultaneously; changes in one part of the system that are not balanced with changes in other parts can be counterproductive.

The ideal system is one in which the goals of owners and the agents who represent them are aligned with those of corporations and of society as a whole. A better alignment of goals would be the most important step forward and would itself lead to beneficial changes elsewhere. We need a system in which more and different information is used to guide valuation and investment decisions. More aligned goals and better information must drive a process by which constructive intervention occurs at all levels when behavior departs from the long-term best interests of the corporation. Finally, the system must be modified to better align private shareholder returns and those of the economy as a whole. What is needed, in many respects, is a reexamination of our entire capital market paradigm.

A more supportive macroeconomic environment, which provides the context in which all corporate investment takes place, will provide a foundation for the other systemic changes needed. Increasing the stability of this environment and enlarging the pool of savings will reduce risk premiums and lower the cost of capital.

Beyond the macro environment, reform is needed in the broad areas of (1) ownership, (2) goals, and (3) information. In the following sections, we will translate these areas into specific directions with implications for policymakers, corporate management, and institutional investors.

The directions for policy reform rest on principles that differ markedly from those that have defined the regulatory framework of the traditional U.S. system of capital allocation. Rather than avoid abuses by restricting the activities of the largest capital providers and large corporate owners, corporate ownership can be broadened while the goals of capital providers, corporations, managers, employees, and society are better aligned. Capital providers become knowledgeable and constructive participants rather than adversaries. Abuses can be prevented by modifying incentives and eliminating unneeded regulatory guidelines rather than resorting to regulatory constraints that carry unintended consequences.

Stated most boldly, our research suggests the need to reexamine much of what constitutes the U.S. system of management, with its extreme approach to managing decentralization, its limited flow of information, its heavy use of certain types of incentive compensation systems, and its reliance on financial control and quantitative capital budgeting processes.

While the U.S. system is partly the result of regulation, there are positive steps that can be taken by institutions without the need for public policy changes. First and foremost, institutions must begin to understand why managements view them as adversaries. They must understand the subtle consequences of their monitoring and valuation practices on corporate investment behavior. They must also recognize that greater influence over management will come only at the price of less flexibility, less rapid trading, and the need for greater knowledge of and concern with company fundamentals.

1. Expand True Ownership throughout the System

The current concept of ownership in the U.S. system is too limited and restricted largely to outside shareholders. While outside owners should be encouraged to hold larger stakes and to take a more active and constructive role in companies, ownership should also be expanded to include directors, managers, employees, and even customers and suppliers. Expanded ownership will foster commonality of interest and help make investors and managers more aware of the value of investment spillovers that strengthen firms, benefit employees, and enhance the economy as a whole. A number of policy changes could shift the pattern of wnership to favor appropriate investment and better address externalities, for example:

• Remove restrictions on share ownership.
  The regulations that artificially restrict the ability of investors to hold significant corporate stakes should be reexamined, as should restrictions on joint ownership of debt and equity. Limits on significant holdings should be eased, diversification rules relaxed, and fiduciary guidelines governing institutional investors modified to better reflect the true risk to owners (it can be shown, for example, that the value of diversification can be gained by holding relatively few stocks, certainly many fewer than the 100, 500, or 2,000 that are now common).|8
• Lower tax barriers to holding significant private ownership stakes.
  Estate tax laws need to be reexamined to strike a better balance between the need to equitably collect estate taxes and the ability of significant private owners to maintain their ownership stakes. Ways of structuring estate tax rules to allow their payment over some extended period, where illiquid or hard-to-value assets are involved, would be desirable.
• Seek long-term owners and give them a direct voice in governance.
  Perhaps the most basic weakness in the U.S. system is transient ownership, in which institutional agents are drawn to current earnings, unwilling to invest in understanding the fundamental prospects of companies, and unable and unwilling to work with companies to build long-term earning power. The long-term interests of companies would be best served by having a smaller number of long-term or near permanent owners, whose goals are better aligned with those of the corporation. This does not necessarily mean taking the company private but could involve a hybrid structure of "private" and public ownership. Ideally, the controlling stake would be in the hands of a relatively few long-term owners,
  though shares were publicly traded to allow access to the public markets when
  conditions were favorable. Some owners might represent syndicates of other, smaller, long-term owners.

  In return for a commitment to long-term ownership and to becoming fully informed about the company must come a restructuring of the role of owners in governance. Long-term owners must have insider status, full access to information, influence with management, and seats on the board. The board, consisting primarily of owners along with suppliers, customers, and others highly knowledgeable about the firm's business, should be both influential and informed.

• Nominate significant owners, customers, suppliers, employees, and community representatives to the board of directors.
  As described earlier, such directors are more likely to have the company's long-term interest at heart and to encourage management to make the investments required to ensure long-term competitive position.
• Encourage long-term employee ownership.
  Ownership by employees is desirable for a variety of reasons, provided that employee owners are long-term rather than transient owners. Employees' goals as owners will tend to be aligned with the long-term health of the corporation, and employee ownership is likely to internalize some of the externalities in investment choices by heightening the pressures for investment in human capital. Taxes and other regulations should facilitate and encourage employee stock ownership. Currently, companies can offer shares to employees at a discount of up to 15 percent of market price with no earnings impact, and such incentives should be maintained or increased. Rules that allow low-cost issuance of new shares for employee ownership should be enacted. To qualify for such fiscal advantages, however, employee stock ownership plans should be required to restrict the sale of affected employee shares for five years except in cases of genuine hardship and to limit the proportion of holdings that can be sold in a given year.

2. Better Align the Goals of Capital Providers, Corporations, Directors,
Managers, Employees, Customers, Suppliers, and Society

More ownership per se will not be sufficient if the goals of owners, corporations, and others are not aligned with each other and with the maximization of long-term corporate value. It is possible to create a system of incentives and to alter rules in a way that helps align the goals of all constituencies. Since the goals of agents will inevitably reflect those of owners, we must address owner goals directly.

The single most powerful and practical tool for modifying owner/agent goals is establishing a significant incentive for making long-term investments in corporate equity. The governing principle behind this proposal is to fundamentally change the system by changing the concept of ownership and the approach to valuing companies, while at the same time encouraging the form of investment where the externalities are greatest.

The current debate over capital gains, focused on encouraging investment and raising short-term tax revenue, misses the point. Currently, ordinary income and capital gains are treated as equivalent, and the debate over capital gains incentives is stalled. Our proposal is to restrict the incentive to equity investments in operating companies. The incentive would not apply to capital gains from bond appreciation or real estate appreciation, from investments in nonoperating companies holding real estate, or from other financial assets.|9

The long-term equity investment incentive would require a minimum five-year holding period, with lower tax rates for even longer holdings. Ideally, capital gains would also be indexed for inflation. The equity investment incentive would be prospective, applying only to new investments and new gains.

Other mechanisms for affecting the goals of owners and lenders include extension of the long-term equity investment incentive to currently untaxed investors such as pension funds and eliminating restrictions on joint ownership of debt and equity.|10 It would also be desirable to reduce the extent of explicit and implicit subsidies for investment in real estate. The positive externalities involved in real estate investment do not seem to justify the disproportionate level of incentives compared to investments in corporate equity, R&D, and training.

Shifts in corporate goals will need to accompany regulatory changes such as those just suggested if they are to be effective. Existing corporation law identifies long-term shareholder value as the appropriate corporate goal. Stakeholder laws, which explicitly provide for the interests of other parties such as employees or the local community in corporate goals, are a less-than-ideal solution to this problem. Stakeholder interests are hard to clearly define, prone to overemphasis, and subject to wide discretion. The granting of director discretion to consider stakeholders' interest (implicitly, at the expense of shareholders) without those stakeholders for corporate performance having responsibility serves to further isolate boards from owners. A far better solution is to explicitly identify long-term shareholder value as the corporate goal and shift the burden of proof on directors and managers to explain to owners or other major constituents why decisions build long-term value. This would be accomplished by modification of corporation laws designed explicitly to reduce the threat of litigation. The interest of
shareholders, employees, the local community, and other stakeholders is far closer, if not identical, in the long run than in the short run. Improved aligning of the interests of owners and corporations, combined with the other changes we have suggested, can go a long way toward properly addressing the problem that stakeholder laws reflect.

Improvements in the internal capital allocation and investment monitoring process must primarily be the responsibility of management. However, some incentives designed to better align private and social returns on selected forms of investment would be desirable. In particular, regulation can provide investment incentives for R&D and training. Such forms of investment are particularly subject to externalities that make the optimal private rate of spending less than for society. Making the existing R&D tax credit permanent and creating a parallel tax incentive for training investments should take on high priority.

The need for traditional investment tax credits, on the other hand, is more problematic. Investment tax credits are expensive and are subject to political distortions. The U.S. system of capital allocation is less biased against investment in hard assets, and the externalities involved in such investments seem more limited. Public resources would seem better spent in funding the long-term equity incentive, which will shift the entire valuation system to encourage investment and the funding of permanent incentives for R&D and training.

U.S. compensation systems need to move in the direction of linking pay more closely with long-term company prosperity and to actions that improve it. Performance awards should be based not only on rate of return for the year but on the extent to which the company's competitive position has improved (or weakened) based on objective measures. Stock options should be modified to include long (more than five-year) vesting periods and constraints on the number of shares that can be liquidated at one time. Otherwise, even some well-meaning managers will manage the stock price rather than build the company.

Another way to align the goals of management with the long-term health of the corporation is to move away from unrelated diversification. Unrelated or loosely related diversification not only wastes capital but also exacerbates the management biases that we have described. Concentrating on one or a few core fields and investing heavily to achieve a unique position in each of them is the only way to build highly competitive companies.

Institutional investors are a special case in that, as we have seen, they control the major part of the external capital market in the United States. The U.S. system of capital allocation creates perverse outcomes for institutional investors, especially pension funds. Such institutions should be the ideal long-term investors. Instead, we have the paradoxical situation in which many institutions, especially pension funds, are entrusted with funds for extremely long periods, yet trade actively. Institutions incur substantial transactions costs, yet most underperform the market. Many have been driven to index funds. Because of the widespread holdings of institutional agents, a gain on the sale of one stock comes as a loss on another stock. Thus, the only way for institutions to improve their overall performance is to improve the earning power of U.S. companies, which will increase their aggregate value. Society also has a large stake in creating a system where the "winning" institutions are those that are better at rewarding and directing capital to companies with the best long-term prospects rather than those whose earnings will improve next year.

Institutions can take the lead in moving toward the following changes.

• Increase the size of stakes
• Reduce turnover and transactions costs
• More carefully select companies based on fundamental earning power
• Encourage changes in agent measurement and evaluation systems to reflect long-term investment performance
• Transform interactions with management to productive, advisory discussions. Create special funds to test these new investment approaches
• Support systemic public policy changes

Institutions might find it easier to make such changes if, as a first step, they were to consider creating special funds earmarked for long-term investing in significant stakes in companies who agree to give major owners a greater influence in management. Given existing rules, establishing such funds might require agreements modifying the funds' fiduciary responsibilities. Such funds would be structured to provide liquidity without the need to sell positions through mechanisms by which existing investors or new investors could buy (at the current market price) the holdings of those investors needing to generate funds.

There are positive changes underway in some of these areas that can be built upon. Even CalPERS, for example, is attempting to reduce its number of portfolio companies, from over 2,000 to between 700 and 800 (White 1991; Lorsch and MacIver 1991). The more institutions that modify their approach to investing, the more the fruits of a shift to committed long-term investing will become a self-fulfilling prophecy, because enough funds will seek out the most competitive companies to drive market prices. Current earnings will begin to have less and less of an impact on share prices. Stock prices could become less volatile. The transactions cost savings alone of this approach will provide a head start in registering better investment performance.

3. Improve the Information Used in Decision Making

Even if goals are better aligned, the quality of information used to allocate capital throughout the system will affect investment choices. We must expand access by both investors and managers to information that better reflects true corporate performance. While some of the needed changes in this area can be initiated via regulation, much of it will depend upon corporations recognizing and correcting counterproductive organizational structures. For example, the mode of managing decentralization in place in many U.S. companies must be overhauled to provide greater information flow and to better align it with the imperatives of competitive advantage. The problem is not that decentralization is bad per se but that it has been inappropriately managed, partly because of excessive corporate diversification.

What is needed instead is a system that recognizes strategically distinct businesses as the proper unit of management but manages them differently. Senior managers at the corporate level must have a substantive understanding of both the technology and the industry. Top management must be directly and personally involved in all significant decisions, especially investments. There must be extensive consultation and coordination with employees and among related business units. Information must be exchanged both vertically and horizontally. Opportunities to share functions and expertise among units should be pursued aggressively because they leverage tangible and intangible investments. Incentives must be changed to promote the internalization of such spillovers within the company.

Such a system would shift measurement and control away from solely financial results, raise senior management confidence in understanding complex investment choices, and better capture complementarities among discrete investment options. A new philosophy of management control must be instituted, based as much or more on the company's extended balance sheet as on its income statement. A company's extended balance sheet measures the assets that constitute its competitive position. Financial measures of investment returns should be combined with measures of competitive position such as market share, customer satisfaction, and technological capabilities. A more appropriate control system must include the following elements.

• Broader definition of assets
• Measurement of asset quality and productivity in addition to quantity
• Relative instead of absolute measures.
• Move to universal investment budgeting
• Evaluation of investment programs instead of discrete projects
• Unified treatment of all forms of investment
• Separate the determination of required asset position from evaluation of the means of achieving it.

Other constructive changes in corporate behavior would include the following points.

Modify accounting rules so that earnings better reflect corporate performance.
The accounting profession, led by the SEC if necessary, should create new standards for accounting for intangible assets. R&D and other long-term intangible investments should be capitalized under rules that require disclosed write-offs of worthless investments as part of the auditing process. While there will inevitably be some abuses, as there are under the current system, the benefits in terms of improved valuation and the resulting shifts in corporate behavior should outweigh them.

Expand public disclosure to reduce the cost of assessing true corporate value.
Accounting earnings, even if modified to better account for investments in intangibles, still fail to provide a measure of true long-term corporate value. Value resides in a company's activities, its stock of scientific and technical knowledge, its skill bas