From YalePressWiki

[edit] Chapter 8: The Care and Maintenance of Entrepreneurial Capitalism

Success breeds complacency. Complacency breeds failure. Only the paranoid survive.
—attributed to Dr. Andrew S. Grove, cofounder of Intel Corporation

For roughly a century, the United States has been the leading economic power in the world. Part of the reason for its success is that the United States has been the quintessential exemplar of a mixture of entrepreneurial and big-firm capitalism. But can Americans, and others in the world who are now emulating the U.S. model, safely assume that the American economy will continue to be as successful as it has been in the past? Or is the United States—like other once-great civilizations (ancient Rome and Greece come to mind)—doomed in the foreseeable future to fall or at least stagnate? We make no claim to be prophets, but we believe that while there is much to be optimistic about, there also are dangers ahead. In this chapter, we will celebrate the former and call attention to the latter in the hope that current and future policy makers will take steps to help us avoid the fate to which the dangers could condemn us.

Put in a nutshell, we will suggest in this chapter that one of the most immediate perils facing the U.S. economy is the possibility of transforming into a much less entrepreneurial big-firm regime, one characterized by ossification, limited incentives, and a paucity of breakthrough inventions. There is no simple formula for preventing such an outcome, but the analysis of this book suggests the importance of two key principles: provide incentives for productive entrepreneurship and discourage diversion of entrepreneurial talent into unproductive or destructive sources of wealth. In this closing chapter, we will provide a number of suggestions about how best to maintain the critical balance of big-firm and entrepreneurial capitalism, in part by simply bringing together the observations of the two preceding chapters.

One cannot simply dismiss the big-firm regime peril. For one thing, the U.S. economy has not always had such a nice blend of big-firm and entrepreneurial capitalism. In the 1950s and 1960s—the heyday of Big Auto, Big Steel, and Ma Bell (the old AT&T telephone system)—the economy was much closer to a regime of big-firm capitalism. Then came the oil price shock of 1973–74, the years of stagflation, and two decades of disappointing productivity growth. A resurgence of entrepreneurial innovation— largely in information technology and communications—coupled with more intense foreign competition (which forced the older big firms to become vastly more efficient and to improve the quality of their products) helped to turn the U.S. economy around. But, as the saying goes, nothing lasts forever. It is conceivable that the U.S. economy might revert to a more strictly big-firm regime, and it is wishful thinking to believe that this pattern could not reemerge.

Indeed, one farsighted economist, the late Mancur Olson, argued that something like this is likely to be the destiny of all economies, especially those in democratic societies (Olson, 1982). As economies age, Olson asserted, special-interest groups grow in number and power; as this happens, it becomes more likely that they will come into conflict. Like physical objects subject to Newtonian laws, calls for action by some special-interest groups meet with counterreactions from others, all aimed at thwarting each other’s ambitions. Too often, the result can be paralysis or “rentseeking” of the worst sort with regulations and policies that benefit particular groups without conferring benefits on, and even detracting from, the general welfare. The proliferation of trade associations and lobbyists— which was apparent when Olson wrote his book in the early 1980s and is even more so now—is powerful confirmation of his insight.^[1] A reversion to big-firm capitalism would threaten analogous effects, leading to interest- group paralysis via a slowing of the rate of radical innovation.

In this chapter we will focus on the United States, in particular considering what it can do to maintain its unusual blend of big-firm and entrepreneurial capitalism in the future. We direct our attention to the United States for three reasons. First, it is the economy we know best. Second, we suspect that many of our readers are Americans. And third, even for readers outside the United States, we believe there is much to learn from the American experience.

We will follow the structure of earlier chapters, where we laid out four broad preconditions for the right of blend of entrepreneurial and big-firm capitalism and four types of economic-health-sustaining measures we considered critical. Here, we will assess current trends relating to three of the most relevant of these requirements to the U.S. economy. In some cases, there is little cause for concern; in others cases, there is plenty cause for worry, and where that is true, we will offer some possible solutions. But lest readers think we are concluding this book by wringing our hands, we begin with the good news: evidence of what is right about the American economy.

[edit] The Productivity Miracle, So Far

Look through the business media and you’ll find a plethora of statistics— about stock prices, the inflation rate, the growth rate of the economy over the past quarter or year, and the current unemployment rate. But in the long run, only one economic statistic really matters: the growth of productivity, which measures increases in output for given inputs. Because the most important input, judged by its share of national income and thus cost for generating goods and services, is labor, many economists focus much of their attention on labor productivity. We do the same here.

Labor productivity, by definition, measures output generated per unit of labor input, typically per hour. Again, by definition, the level of labor productivity at any one time reflects the average standard of living of the residents of any economy, while the growth rate of labor productivity measures the rate at which that average living standard improves.^[2] This is not to imply that if a worker’s productivity rises by 7 percent, she deserves all the credit. Improved technology, the availability of more and better raw materials, and other such developments make a critical contribution, and there is good reason to argue that in recent centuries innovation has been the primary source of productivity growth. Thus, while it is relatively easy to measure productivity and its growth, full determination of what was responsible for that growth is exceedingly difficult.^[3]

Figure 6. Index of U.S. Productivity (Nonfarm Business), Real Output per Hour, 1948–2005. Source: U.S. Bureau of Labor Statistics, Major Sector Productivity and Costs Index, 2006. Available at http://data.bls.gov/.

We do know, however, that productivity in the United States and other wealthy economies, despite accelerations and slowdowns, has been growing virtually without interruption for the past two centuries (see figure 6). The rate of growth has been beset by ups and downs. In the United States, for the first quarter-century after the World War II—that is, between 1948 and 1973—productivity grew at an average rate of 2.8 percent per year, according to the U.S. Bureau of Labor Statistics. Between 1973 and 1995, it grew far more slowly: about 1.4 percent annually. But between 1995 and 2002, it recovered to 2.8 percent per year and through 2005 advanced at an even higher rate (a little more than 3 percent). The difference between a 1.4 percent and a 2.8 percent rise per year in our incomes may not seem like much, but the difference increases every year because it is compounded like interest in a bank account. At a steady growth rate of 1.4 percent per year, our descendants one hundred years from now will receive incomes on average four times as large as ours in purchasing power. But if, instead, the steady rise continues at 2.8 percent a year, our descendants’ purchasing power will have risen to be sixteen times as large as ours. In the rate of growth of productivity, what appear to be minor differences soon matter very much indeed.

Many economists believe the resurgence in productivity in the 1990s was to a considerable degree attributable to what has been labeled IT (information technology), including computers, computer software, the Internet, and the like. There has been spectacular growth in productivity in the IT sector itself. For example, data collected by the U.S. Bureau of Labor Statistics show that labor productivity (output per hour) in the computer and electronic products manufacturing industry rose at an average annual rate of 15 percent between 1993 and 2003, compared to an average of 4 percent for manufacturing as a whole (U.S. Bureau of Labor Statistics, 2006; and Schweitzer and Zaman, 2006). This extraordinary productivity growth rate in computer manufacturing has been driven by advances in computer technology and the technology employed in the production processes of the computer industry. The rapid growth rate also illustrates “Moore’s law,” the prediction by Gordon Moore (cofounder of Intel, the world’s largest computer chip manufacturer) that the number of transistors on a chip will double every eighteen months or so.

Productivity in the rest of the economy that uses IT has also benefited from the IT miracle. As was the case with the introduction of electric power for industrial and household use, it took decades before IT had much of an influence on productivity growth in other sectors of the economy. But once other industries learned to make use of it, the results were substantial, and there is reason to ascribe a significant part of the rise in growth of labor productivity in the 1990s to the expanded role and capacity of IT.^[4]

The most recent data indicate that labor productivity in the United States has advanced at a more rapid rate since 2000: at an extraordinary 3.5 percent through 2005. Economists are deeply divided about whether this high rate can be sustained in the future, and if not, how much of a fallback is likely. The most likely outcome, in our view, is some falloff in productivity growth because, eventually, the diffusion of productivity-enhancing information technology will slow down, and, indeed, official government projections reflect that. The Congressional Budget Office and the Social Security Administration, for example, project long-run future productivity growth to be little more than 2 percent—above the disappointing 1.4 percent average of 1973–95 but well below the 3 percent growth rate of the past decade. History suggests that periods of extraordinarily rapid or especially slow productivity growth do not last very long. Indeed, after surging at close to a 4 percent annual clip since 2001, labor productivity grew at less than 3 percent in 2005.

There is much at stake here for future generations. The difference, in terms of the long-run earnings consequences, between an annual rate of productivity growth of, say, 2 percent and 3 percent is enormous, as indicated in table 17.

Which of these growth rates the future will bring is very likely to depend on the nature of the next big innovation breakthrough(s)—that is, whether or not it or they will be something as important as the Internet has been. Some breakthroughs may provide huge social benefits but not show up that way in the economic statistics; major advances in biotechnology or other health endeavors are likely examples. These will result in a longer lifespan for many of us, but the benefits of added longevity will not show up in higher GDP. On the contrary, as the population ages, the burden of current entitlement programs on current and future workers will grow, leading to higher taxes and/or higher interest rates that result from added government borrowing, both of which are likely to dampen growth. Other breakthroughs, such as nanotechnology or some combination of new products and production techniques that cannot be foreseen now will show up in measured GDP. If our argument in this book is correct, these radical breakthroughs are more likely to come from smaller and newer firms, although larger, established firms will be essential to mass production and increased reliability and capacity of whatever innovations the smaller firms come up with.

Continued economic growth is not just important to satisfy Americans’

competitive desire to remain “number one.” Growth is imperative if we want to continue to improve the quality of our lives along many dimensions. Growth is also critical in light of the rapid aging of our population, a trend that is even more apparent in Japan and Western Europe. An aging population will impose large and increasing obligations on our government to make good on the promises of the nation’s entitlement programs— Social Security and, perhaps more important, Medicare—at a time when the country faces many other urgent priorities, including increased spending for national security at home and abroad, to defend against terrorism, as well as pressures to expand health insurance for the nearly 50 million Americans under the age of sixty-five who do not have it. Under these circumstances, anything that can be done to enhance the underlying rate of technological change, and hence economic growth, will be welcome.

For all these reasons, it will be critical for the United States to maintain and, ideally, improve upon the institutions that satisfy the four preconditions outlined in chapter 5 for ensuring the right blend of entrepreneurial and big-firm capitalism. Here, and in the rest of this chapter, it will be convenient to combine the four preconditions into three:

1. Adequate incentives for productive entrepreneurship, including appropriate rewards and adequate security of those rewards and of the earnings processes, must be maintained and ideally strengthened;
2. Disincentives for unproductive entrepreneurship must be restricted and ideally eliminated; and
3. Continued rivalry among and innovation by large firms must be ensured.

[edit] Incentives for Productive Entrepreneurship

The first of our three growth preconditions is the body of incentives for people to become entrepreneurs and to employ productive means of pursuing their personal goals, presumably some combination of wealth, power, and prestige. For this purpose, in simplest terms, society must offer some of its highest rewards to the most successful productive entrepreneurs. In particular, the rules and institutions must remove impediments to such activities, they must provide or at least not interfere with financial rewards for success, and they must provide security for these rewards, offering some degree of assurance that they cannot simply be taken away.

The last of these issues is, in general, no longer a matter of serious contention as it used to be in earlier periods of history in some noncapitalist societies, when kings or robber barons or warlords could simply expropriate the earnings from productive activity. Today, laws defining property rights and the penalties for fraud or theft provide reasonable security for the earnings of the productive entrepreneur, at least in economically advanced nations. Only one element in the set of institutions that ensures the requisite security of these asset accumulations—the patent system—requires more extensive discussion below, and it makes most sense to do so in the United States context. Although patents are necessary to provide incentives for productive entrepreneurship, the United States system in recent years appears to have become too protective, which can not only discourage new entrepreneurs from creating new markets but also insulate existing firms from the hot breath of competition.

On most other scores, the United States maintains a good record for encouraging productive entrepreneurship on all the requisite fronts. It has well-developed systems of property rights, and in recent years it has cut back marginal income tax rates as well as some regulations that hamper the task of the productive entrepreneur. Nonetheless, the income tax system will be under increasing pressure as the U.S. population ages and funds are needed to fund entitlement programs, such as pensions and health care. The corporate income tax, assuming it remains in place, can be restructured to strengthen the incentives for growth. Recent new regulations related to financial reporting and corporate governance—enacted after the financial reporting scandals earlier in this decade—now appear to be unduly burdensome for new enterprises. We will take up each of these subjects in the material that follows.

[edit] Removing Impediments to the Launch of Productive Enterprise

The most obvious step to make productive entrepreneurship attractive is to remove any handicaps to the creation of new firms. This is more difficult than it may seem. In many parts of the world, the bureaucratic impediments to firm-creation are substantial. In the United States, the technical, legal, and institutional costs associated with starting a business, fortunately, remain low, as the World Bank’s Doing Business reports have consistently documented. The Internet is likely to further reduce these costs as more states and localities make it possible to obtain necessary licenses and approvals without having physically to appear at a government office. The same will be increasingly true in other countries as they embrace the Internet for the same purpose.

More broadly, the technological advances in the information technology (IT) and communications industries have substantially reduced the capital costs associated with entering some IT- and many Internet-related businesses. For retailers and middlemen, the costs may amount to as little as several thousand dollars, representing the cost of acquiring a computer, some related equipment (such as a printer, scanner, and fax machine), and some off-the-shelf Web site preparation software.^[5] Indeed, with start-up costs now so low, many Internet businesses no longer need venture capital to finance their inauguration because they can attain positive cash flows in a relatively short period of time. This is clearly a beneficial development. The Internet is becoming a powerful engine that can wring out costs and increase productivity throughout the economy (Litan and Rivlin, 2000).

Yet here, as elsewhere, there are grounds for concern about the more distant future arising from the increasing tendency, at least in the United States, of many start-up companies to sell out to larger firms rather than to “go public,” which was a common exit strategy for founders of and earlystage investors in high-tech companies in the 1990s. Indeed, a substantial number of larger firms—Microsoft, Cisco and Intel, among others—effectively have been outsourcing much of their R&D to the entrepreneurial marketplace, waiting for innovative companies to prove themselves and then snapping them up, or at least purchasing stock in them, much as the venture capitalists did for many high-tech companies in the 1990s. This has left some venture capitalists searching for other investment opportunities.

Our concern here is not for the welfare of the venture capitalists, but we do worry that small, young innovative companies built on or around radical and disruptive technologies will not be able to attain their true potential capacities—nor will their innovations—if they are regularly absorbed by larger, more bureaucratic firms, which may not have the same entrepreneurial spirit or culture. True, this has gone on in the recent past to a substantial degree, and so far the depleted ranks of innovative firms appear to have been replaced by entrants who grasp the opportunities provided by the departure of their successful predecessors. But this process remains a matter to be watched, and in the future policy adjustments may be required to keep the pipeline of innovative start-ups full.

One other worrisome public policy development that may affect the launch of future innovative companies is the unintended effect of the recent tightening of the bankruptcy law. Congress amended the U.S. bankruptcy laws in 2005 unaware that as many as 20 percent of all personal bankruptcies may, in fact, be business-related (Lawless and Warren, 2005). Since many entrepreneurs begin by bankrolling their activities with credit card charges, the modifications of the bankruptcy law, which force those who declare bankruptcy to repay more of their debts (seemingly a good thing), may unintentionally discourage the formation of new enterprises. Effective bankruptcy laws are important because the more difficult it is to exit from a business, the less likely it is that innovative entrepreneurs will take the risk of getting started in the first place. We must not forget that impediments to the exit of an unsuccessful firm can be, in effect, the equivalent of an increase in the cost of entry.

[edit] Taxation Policy

It is common in books that, like this volume, focus on forces the authors believe to be crucial for growth to recommend that governments provide some added incentives to strengthen those forces. The incentives perhaps most frequently proposed are tax deductions, tax credits, or lower tax rates for activities that contribute to economic expansion. In fact, federal policy makers in the United States in recent years have provided substantial tax incentives for all kinds of activity, which account for the complexity of the current tax code. In 2001 and 2003, President George W. Bush proposed and the United States Congress approved two tax bills, substantially lowering marginal income tax rates for individuals, especially for upper-income taxpayers, and (through 2007) establishing a 15 percent rate on payment of corporate dividends. In combination, the two tax bills lowered the federal income tax share of GDP to roughly 17 percent, from 20 percent in 2000. In November 2005, a special advisory panel on tax reform appointed by President Bush outlined broad recommendations to simplify the tax code by disallowing a number of personal deductions (notably, the one for state and local income taxes) and, as a trade-off, reducing the marginal income tax rates for individuals still further. The panel also offered recommendations related to the corporate income tax to enhance the incentives for investment (by proposing the immediate expensing of all new investment, thus shielding it from taxation, but denying any deduction for interest on borrowing). One study has argued that by reducing some of the distortions in the current tax system, the simplification elements in the panel’s tax recommendations should benefit entrepreneurs (Bruce and Gurley-Calvez, 2006).

As important as simplicity is in structuring a tax system, the central objective of any tax system is sufficiency—that is, collection of enough revenues to support government expenditures. For this purpose, the Bush tax panel’s mission was of no value because it was instructed only to examine proposals that would not affect total revenue collected. Yet this instruction ignored some clear budget mathematics. With the impending retirement of a generation of baby boomers, coupled with near-certain continuation of rapidly rising health care costs (albeit while research and technical progress improve its quality), federal spending on the government’s three main entitlement programs—Social Security, Medicare, and Medicaid—is projected to rise rapidly. In particular, unless the benefit structures are changed, federal spending on these programs, which totaled a bit more than 8 percent of GDP in 2005, is projected to reach nearly 14 percent by 2025 and to hit 19 percent by 2050.^[6] Because other federal spending, including interest, is likely to total at least 12 percent of GDP throughout this period (its level in 2005), it is clear that the tax share of GDP cannot remain at 17 percent without having an explosion of the deficit as a share of GDP. At some point, it is inevitable that a future Congress and president must agree on a budget package that cuts spending growth in the entitlement program and also raises taxes.

Of course, no one likes a tax increase, and selling one to a public accustomed to the lower marginal income tax rates implemented through the 2001 tax bill will be extremely difficult. From an economic perspective, it will be important that any tax increase do the least harm to long-term economic growth. What kind of tax increase best meets that test?

In principle, additional revenue raised through taxes on consumption rather than higher income tax rates should do the least economic harm because consumptions taxes should encourage private saving, which in turn should reduce interest rates and thereby increase investment (which could conceivably lead to higher growth in the long run though slower growth in the short run due to the depressing effect of consumption taxes on aggregate demand). In contrast, raising income tax rates entails some risk of discouraging work effort, although an across-the-board increase in income tax rates for everyone (employees and business owners alike) would not necessarily discourage entrepreneurial activity in particular.^[7]

Consumption taxes have their drawbacks, however. For one thing, it is difficult to design a tax on consumption that is progressive in impact, namely, one under which taxpayers with low incomes pay lesser amounts relative to their incomes than those with higher incomes. In addition, and more pertinent to our subject at hand, certain kinds of consumption taxes—a value-added tax, for example—could actually hurt entrepreneurs by requiring them to pay taxes on inputs of production before they earn revenues on the sales of their products and services (and thus become eligible for rebates on those input taxes). Entrepreneurs may also suffer an adverse impact on their cash flow from a straight sales tax on certain inputs.

Accordingly, a more progressive way of raising revenue without discouraging work effort or entrepreneurship would be to “broaden the income tax base”—that is, to keep marginal income tax rates where they are but cut back on deductions and exemptions. This may be the ideal economic outcome, but a broaden-the-base approach may be the least politically palatable way of raising additional revenue since it would target specific, identifiable groups—those that now benefit from the deductions (the home-building industry, state and local government officials, and charitable organizations, to name just a few)—rather than impose burdens across the board.

Clearly, it is not to our comparative advantage to outline here the tax package that is least politically damaging and that least hurts growth and entrepreneurship. All we can do is highlight the trade-offs involved while underscoring that the alternative, doing nothing—that is, taking no action to raise revenue, given the certain continuation of Social Security, Medicare, and Medicaid in something resembling their current form—is worse than any of the revenue-enhancing options, for entrepreneurs and for all Americans, in the long run.

[edit] Financial Reporting Requirements and Sarbanes-Oxley

Legal requirements and the effort and costs needed to meet them can hinder entrepreneurial development. These costs may not be imposed when the firm is first created, but if they dim the prospects for profitable expansion later on (should the new firm should prove successful), they could constitute a disincentive to productive entrepreneurial activity.

A significant illustration is provided by the policies recently adopted in an effort to prevent the financial misconduct associated with Enron, WorldCom, and a number of other high-flying companies, whose failures led to the 2002 Sarbanes-Oxley Act (SOX) and related changes in listing requirements of the major stock exchanges. Among the reforms are requirements for: a majority of public company boards to be made up of independent directors (rather than in-house corporate managers); public companies to undergo extensive annual internal and external audits of their internal controls (the mechanisms to assure that corporate funds are not being misspent); and the remaining four large public accounting firms (as well as any smaller ones that audit public companies) to be subject to the oversight of a new regulatory body, the Public Company Accounting Oversight Board (PCAOB).

Of particular interest to us are the extensive auditing and independent director requirements. The open question is whether the costs of such measures deliver sufficient benefits—that is, a reduction in misconduct beyond what was likely with the previously existing criminal penalties for those involved in corporate financial scandals. At the very least, the additional costs of “going public”—that is, of a firm that formerly was privately held inviting the public to acquire part of its ownership via the purchase of company stock—may be contributing to the tendency of successful startups to avoid public offerings and to sell out to large established companies instead. In addition, there are anecdotal reports that whether or not SOX deters young companies from going public, it may be discouraging their founders from staying with the enterprises after they have done so. More public companies are also “going private,” perhaps to avoid the regulatory burdens associated with SOX.

It is difficult to know how significant any or all of these effects are and will be in the future, but as the saying goes, “where there’s smoke, there’s fire.” Although we are not ready to conclude that the costs of SOX have outweighed the benefits—it is conceivable that in the aggregate the reverse has been true—there seems to be sufficient “smoke” surrounding the impacts of SOX to warrant further examination of the law by academic researchers and policy makers to determine whether it or the regulations implementing the law might be modified to make it more cost-effective.

One good place to start would be to refine the Section 404 rules to make audits of internal controls more risk-based than they currently appear to be. Such a review should apply to the implementation of SOXrelated rules for smaller and larger public firms alike. At this writing, the Securities and Exchange Commission (SEC) appears to be engaged in just such an activity and in December 2006 proposed clarifications to accomplish this.^[8]

Another, more ambitious idea is for the SEC to specify different levels of “404 intensity” and then let shareholders choose which ones they want. Thus shareholders who are suspicious, wary of, or concerned about current management may vote to have the most intense (and costly) audit of internal controls performed; shareholders more trusting of management would be likely to opt for a less onerous version. Admittedly, it would take some work to flesh out the details of these options, but in principle at least, we suspect there would be a lot less second-guessing about the relative costs and benefits of 404 audits if investors had some say in the matter.

[edit] Legal Protection of Intellectual Property

The legal protection of ideas—in the form of patents, trademarks, copyrights, and trade secrets law—has long been part of the American legal fabric. Patents and copyrights, in particular, are mentioned in the Constitution. Patent rights have their origins in Italy, dating from the fifteenth century.

In principle, so-called intellectual property rights are supposed to encourage inventors and entrepreneurs to engage in activities that generate and propagate innovations. Reality is more complicated. When it comes to promoting entrepreneurship, the protection of intellectual property rights cuts both ways. On one hand, some legal protection surely is warranted to provide incentives for innovation, though the lion’s share of the rewards for innovation accrues to society as a whole, not to the inventor or original entrepreneur.^[9] On the other hand, too much legal protection—in particular, mistaken protection of products or methods of production or service delivery that are not truly novel—can retard innovation and entrepreneurship. Inappropriate or excessively broad legal protection raises barriers to entry by entrepreneurs, discouraging some from developing or promoting new processes or products altogether. Finding the right balance, or threading the needle between these two outcomes, is difficult and yet vitally important.

Institutions created for the protection of intellectual property give rise to a second conflict of purpose. Patents and copyrights, as the means of protecting society’s interests in intellectual products, have two primary objectives. One goal is to ensure that the creators of the property have an opportunity to obtain some reward from their efforts, both as a matter of equity and as an incentive for the expenditure of further creative effort. But the second and apparently rather incompatible goal is ease of access and dissemination to others, to ensure that the benefits of the innovation to society as a whole are as substantial and as widely available as is reasonably feasible.

The conflict between these two goals is widely recognized. The lower the hurdles to accessing intellectual property, the less its creators can hope to charge for its use. If just anyone can make use of new, legally protected ideas, with no impediment whatsoever, the price of access is apt to be driven toward zero. There is a way, however, to reconcile these two goals, at least in principle. Contrary to what one might suppose, patents generally have not served primarily to impede dissemination, but in a wide range of circumstances they have facilitated and encouraged it. To appreciate this, it is useful to take a quick historical detour before returning to how patent law can promote both invention and its disclosure.

The historical oddity is that patents began not so much to reward the creation of new, commercially useful knowledge but rather to promote its transfer from one country to another. As North and Thomas (1973) have shown, so-called letters patent date from the 1300s in England and, as is true today, granted a monopoly to the recipients, for a specified period, over production and sale of the item named in the letter.^[10] Initially these rights were granted not to the creator or inventor of the invention, but to a foreign producer who could steal the idea from his own country and export its use to England. In other words, patents were used to induce technology transfer, not necessarily technology creation. One of the first such letters was awarded to a Flemish weaver for this purpose. In the ensuing years, England encouraged the relocation of many other activities (aside from weaving) from the Continent across the English Channel: mining, metal working, silk manufacturing, ribbon making, and so on. Indeed, of the fifty-five grants of monopoly privilege made under Elizabeth I, twentyone were issued to aliens or naturalized subjects for a variety of products.^[11]

The modern practice of awarding patents primarily to the inventors within a country was adopted into English law in the Statute of Monopolies of 1623 in the wake of parliamentary anger over royal misuse of letters patent to reward royal favorites, and for other purposes having no connection with incentives for generating innovations. Since then, and particularly in recent decades, the voluntary dissemination of patented material has become a major economic activity. More to the point, patent laws around the world since have required holders of patent rights to disclose the technical details that justify the patent. Thus it is that patents, rather than impeding the process, have played a key role in making efficient and voluntary dissemination possible and attractive to the patent owner. Indeed, since at least the latter half of the nineteenth century, the sale or rental of access to intellectual property has become so attractive that it has resulted in the creation of markets dedicated to such transactions with the assistance of professionals who have specialized in the required activities.

In fact, today the sale, licensing, and trading of technology has become a large-scale activity. Arora, Fosfuri, and Gambardella list a sample of “leading deal makers in markets for technology” that includes numerous large companies.^[12] They further report the results of a survey of 133 companies by a British consulting firm, indicating that 77 percent of the companies studied had licensed technology from others, while 62 percent had licensed technology to others. But they also note: “When compared to internal R&D, however, licensing is a fairly modest activity in terms of budgets involved. The survey estimated that expenditures for licensing technology from others amount to 12 percent, 5 percent and 10 percent of the total R&D budgets of North American, European and Japanese respondents, respectively” (Arora, Fosfuri, and Gambardella, 2001, 30–31).

For a number of firms, participation in markets in technology is of critical importance. For example, the sale of access to polypropylene technology has constituted a major activity of the Union Carbide Corporation. IBM has informed one of the present authors (Baumol) that it has a technology exchange contract with every major manufacturer of every significant computer part throughout the world. It is clear that voluntary dissemination is no isolated and unusual phenomenon.

Yet, there is much room for improvement in the operation of patent law, particularly in the United States, where mounting evidence suggests that the balance between too-strict and too-loose patent rules is not being struck as well it could, and thus patents are being awarded for inventions that are “obvious” rather than “novel.” The most infamous example is the decision by the Patent and Trademark Office (PTO) to grant a “business process patent” to Amazon for the “one click” feature on its Web site (“click here to purchase”). But the marked increase in the raw numbers of patents granted strongly suggest that the PTO has become overwhelmed and thus more inclined to give patent applicants the benefit of any doubt. In 2004, for example, the PTO issued 181,000 patents, up from 99,000 in 1990. New applications are running at the rate of 400,000 per year. With the current PTO staff, it would take two years just to clear the backlog let alone process the continuing avalanche of new applications (Orey, 2006).

Professors Adam Jaffe and Josh Lerner argue that the PTO has become more liberal in part because the judges on the specialized appellate court that Congress created in 1982 to hear intellectual property cases have sided overwhelmingly with patent claimants (Jaffe and Lerner, 2004). These judges arguably are behaving much like government bureaucrats who seek to maximize the reach of their agencies’ activities. Another contributing factor is that the PTO is short-staffed and its budget is tied to the fees it collects on the basis of patents granted. As a result, examiners have insufficient time to determine independently whether a patent application is truly novel, and the agency, in general, has a monetary incentive to grant more patents. Whatever the reason for the apparent shift in patent decisions, Jaffe and Lerner are not alone in their critique of the current system. Two prestigious bodies—the National Research Council (Merrill et al., 2004) and the Federal Trade Commission (FTC, 2003)—have reached similar conclusions.

Of course, interests defending strong legal protection of intellectual property—especially the software and entertainment industries—hold a very different view. At a minimum, they favor strict enforcement of existing patent and copyright laws, especially against such new technology as Napster, which allows Internet users to easily copy music and videos. (The courts have been sympathetic to these efforts.) Some support stronger intellectual property laws. There is also a middle view, one that recognizes the potentially sclerotic effect of proliferating patents and copyrights, but which downplays this danger knowing that because few firms can monopolize all the intellectual property necessary to manufacture increasingly complicated products, most will have powerful incentives to cross-license their IP rights.^[13]

Our view is that there are crevices in the system that can stand improvement. Jaffe and Lerner, the National Research Council, and the FTC have each have offered a number of technical suggestions with the objectives of improving “patent quality” (so that patents are granted only for truly new and nonobvious innovations), reducing uncertainty about the likelihood of gaining patent protection, and lowering the costs of obtaining a patent. These are unobjectionable proposals, at least in principle, and the specific suggestions they outline merit serious consideration.

For example, there are calls to permit third parties to contest a patent before it is granted and to give stronger teeth to any post-grant administrative review (for example, by changing the legal standard for overturning a patent from one requiring “clear and convincing evidence” to one based on the “preponderance of the evidence”). Another idea is to enhance the ability of judges or specially appointed masters (rather than juries) to decide the very technical issue of whether a particular invention is novel and nonobvious. There is a growing consensus, even among opponents to the foregoing reforms, that the PTO should be given additional resources to do its job, though any new funding should be decoupled from patent fees (so as not to encourage the PTO to engage in rubber-stamping of applications). Admittedly, to the nonspecialist these recommendations seem like technical fixes, but taken together, they move in the right direction: toward a better system that grants protection to worthy advances while reducing the likelihood that undeserving applications will be approved.^[14]

United States policy makers might also do well to emulate several features of the Japanese patent system, which is designed to encourage dissemination of inventions. These include:

• Awarding of patents to those first to file rather than the party proving that it was the first to invent the item in question. Such a system that encourages early filing would make technical information available sooner than under the current U.S. system. (Such a system is used widely throughout the world, but not in the United States.)
• Allowing others to use a prospectively patentable invention during the period after a patent application has been filed and requiring them simply to pay royalties that are deemed “reasonable.” During this disclosure period, the user is obligated to pay such royalties only if it can be shown that the applicant’s invention has been used “knowingly.” This system clearly adds to the risks applicants face in failing to reach early agreement with prospective users.
• Defining the coverage of a patent in a very narrow sense, so that only the unsanctioned production of items whose technology is extremely similar to the patented item is prohibited. At the same time, the requirement, found in any patent system, that the patented idea be novel, is interpreted relatively loosely in Japan. As a result, rivals who have learned the technology of an innovation during the disclosure period after the patent application, and have used the knowledge to design minor modifications of the original invention, are entitled to apply for patents of their own to cover these variations. This clearly weakens the power of a patent to exclude use by others. In addition, the original inventor may be boxed in by the derivative patents and prevented from using her own invention in her production process for fear of violating imitators’ “improvement patents.”^[15]

Each of these patent provisions puts strong pressure on Japanese innovators to enter into cross-licensing arrangements with rivals. The mere presence of these rules constitutes a threat not only to the success of the initial patent application (via the prospect of direct opposition), but also to the right to use the patent should it be granted but then hemmed in by patented improvements. In Ordover’s words, “The Japanese patent system subordinates the short-term interests of the innovator in the creation of exclusionary rights to the broader policy goals of diffusion of technology” (Ordover, 1991, 48). The Japanese model thus illustrates one way the “rules of the game” can be designed to provide the incentive for managements, particularly large firms with substantial stocks of proprietary innovations, to increase their contribution to the economy’s productive capacity and to its growth.

There was a time, not so long ago, when the idea of reforming the patent system along these lines would have been viewed as politically inconceivable. The forces behind “strong IP” looked too strong to permit any changes that could be depicted as “weakening” IP protection. But one highly visible episode during 2005 and 2006 may have changed the political landscape and set the stage for constructive patent reform in the near to intermediate future. During that time, the inventor of one of the key technologies underlying the widely used Blackberry device (which permits wireless e-mailing), NTP, Inc., came close to obtaining a court injunction that would have shut down the service. The threat was perceived to be so serious that United States government attorneys urged the court to grant employees of the federal government and the congressional and judicial branches an exemption from such an order if it were issued. At the end of the day, the shutdown threat was removed when, at the last minute, the Blackberry manufacturer (Research in Motion, or RIM) settled out of court for more than $600 million. Because of the wide number of Blackberry users, this seemingly arcane lawsuit by a company that engaged in no manufacturing itself—but merely held a portfolio of patents—awakened policy makers and the public to the potentially enormous significance that patent principles and their application can have not only on innovation but on all those who use and depend on it.

[edit] Discouraging Unproductive Entrepreneurship

The second element in the three-pronged program we advocate is reducing or, ideally, eliminating incentives for individuals with entrepreneurial talent to turn their abilities in unproductive directions, which can actually damage a nation’s productivity. Unproductive entrepreneurship can take many forms; for example, enterprising corruption, the formation of organized crime syndicates, or engaging in activities that are legal, such as lobbying legislatures to induce them to adopt laws that bring profits to the lobbyists or their clients. The courts and regulatory agencies provide many opportunities for a clever lawyer-entrepreneur. Relative to other countries, the United States has little problem with corruption, at least as measured by the rankings of Transparency International, a nongovernmental organization that ranked the United States seventeenth out of the 150 on its 2005 corruption perception index.^[16] A larger problem, presaged by the late Mancur Olson (1982), is the mounting collection of interest groups and lobbies, which lead, at best, to inefficiency and, at worst, detract from growth. Examples are not hard to find: unsuccessful efforts to seriously reform large government-sponsored entities (GSEs), such as Fannie Mae and Freddie Mac; lobbying by U.S. farmers to obtain large subsidies (which makes it difficult for the United States to advance trade liberalization); lobbying by the high-tech and entertainment industry to extend intellectual property rights (which, as noted above, may well have gone too far); and rent-seeking litigation, in which firms compete to obtain monopoly licenses or sue their more successful competitors in hopes of obtaining protection from competitive activities. Each of these activities makes use of the innovative efforts of those who represent the parties involved. Resourceful lobbyists can think of new approaches for attaining their objectives, and ingenious litigants can be innovative in their courtroom activities, thereby obtaining large payoffs if their efforts succeed but in the process simply transferring resources from one pocket to another without contributing anything to total output.

It is not easy to solve these problems. It may seem that the lobbying problem can be curtailed by such measures as publicizing monetary contributions to the campaigns of elected officials. But the proliferation of advocacy organizations, which are not subject to disclosure requirements and, perhaps, cannot be under the United States Constitution, has shown how easy it is to get around such rules.

Governments can also adopt rules that limit rent-seeking litigation, although these can be controversial and could be inconsistent with preserving the ability of injured parties to seek redress for legitimate grievances. An example of constructive change is legislation enacted by the United States Congress in 2005 designed to stop “forum shopping” by class-action plaintiffs, that is, the search for friendly state courts in which to press their cases. Under the new legislation, when plaintiffs reside in different states than defendants (which are typically large corporations in class-action cases), the suits must be brought in federal court. A further reform that seems promising is placement of cases involving highly technical subjects— such as medical malpractice cases—in specialized courts, much as patent cases are. This may not be a panacea, however, since these courts, too, can be subject to capture by vested interests (as appears to be the case for the patent intellectual property system [Jaffe and Lerner, 2004]). Nonetheless, some further measures to curtail unproductive entrepreneurship might be implemented, at some point.

[edit] Avoiding Misuse of Antitrust Laws

History is replete with examples of firms that lose out to competitors because of their inefficiency or the inferiority of their products and then seek protection from their rivals, claiming that the competitive actions of their rivals are “unfair,” “predatory,” or “destructive.” There are, of course, cases where one firm does, indeed, adopt measures whose only conceivable purpose is to destroy rivals, as when firms use their deep pockets to adopt prices well below any pertinent costs of their products in the hope that their rivals will be driven into insolvency trying to match them. But there are also examples of what has been called “sham litigation”— lawsuits launched, for example, when a rival’s low costs permit it to adopt very low prices that really are legitimate and serve the public interest. Knowing that the high cost, the energy entailed, and the uncertain outcome of such litigation can induce the defendant firm to surrender and even pay the plaintiff firm an amount sufficient to induce it to withdraw its complaint, firms that otherwise fear they will not succeed on their own merits have incentives to file such suits as a form of insurance against failure.

Indeed, one of the authors of this volume was involved in precisely such a case, in which a group of manufacturers of soft drinks had joined together to form a cooperative enterprise to produce their bottles at about half the cost they had been charged by the then-monopolist bottle manufacturer. That manufacturer communicated with his former customers, telling them that unless they agreed in the future to buy from him exclusively, at whatever price he chose to set, he would launch an antitrust lawsuit against them. The customer firms refused and the bottle manufacturer proceeded to sue. Fortunately, in this case, the judge was made aware of the plaintiff’s undisguised threat, and the lawsuit was thrown out. But such cases do not always end so appropriately, particularly because litigation with exactly the same appearance can be entirely legitimate.

There are a number of steps that can be taken to discourage such enterprising litigation. Under prevailing rules in the United States, if the plaintiff in such a case is victorious, it can expect to be awarded court costs as well as treble damages (see below). But if the defendant wins, and even if the plaintiff’s case is determined to have no merit, the plaintiff is generally not expected to cover the legal costs incurred by the defendant. Clearly, this system can encourage the launching of baseless legal complaints, perhaps even ensuring that the plaintiff has little to lose in the process. One way to correct this problem is to adopt the “English rule” on attorneys’ fees—the loser pays—for some classes of cases, such as commercial litigation, that is, those with commercial interests on both sides. To do it more broadly, however, can make it difficult for legitimately harmed consumers to gain redress of their grievances.

Another idea is to revisit the longstanding provision in U.S. antitrust law that allows successful plaintiffs to collect treble damages if they win. This only adds to the attraction of sham antitrust litigation of the sort under discussion, which can end not only by protecting the plaintiff from the competition of a more efficient rival but also with a substantial financial reward. Treble damages were adopted for two reasons. First, a firm that violates the antitrust laws stands a considerable chance of getting away with it, and if it is in no danger of having to pay its rivals more than the damage its illegitimate activities has cost them, it will proceed with the violation. If so, treble damages may constitute an appropriate deterrent. But the second justification for trebling of damages is more suspect: to enlist “private attorneys general” to uncover and prosecute antitrust abuses. Although some legitimate cases still start this way, the hard work of investigating and prosecuting cases is generally carried out by government officials, with private plaintiffs coming in after the fact to pick up a large damage award. Because plaintiffs would have an incentive to bring their complaints to the government even without trebled damages, we favor the removal of the treble-damage provisions in U.S. antitrust law.

[edit] Reining in Other Unmeritorious Litigation

Lawsuits without merit not only impose needless transaction costs, they can discourage entrepreneurship and growth. This is especially so in unmeritorious class action lawsuits, typically brought on a contingency basis, which can involve damage claims running well into the hundreds of millions, if not billions, of dollars.

A contingency lawsuit, commonly undertaken by increasingly sophisticated well-networked members of the plaintiff’s bar, is one arranged so that the plaintiff pays no attorney’s fees in advance; lawyers are compensated by a substantial share of the award if they win the case. This invites the practice of some lawyers of making their living by seeking out prospective cases, which would otherwise never reach the courts, and inducing prospective clients to serve as plaintiffs, at no financial risk to themselves, representing a much wider class of parties (technically, there are ethics rules to prevent all this, but successful plaintiffs’ attorneys seem to have ways to avoid them).

Class action lawsuits proceed in two stages. First, the court must determine that all class members are “similarly situated”—or damaged by the same cause—so that the case can be treated as a “class” action. In considering whether to make this finding, courts implicitly assume that the defendant is liable, and thus no evidence on that issue can be presented. Second, once a class is certified, the court proceeds to consider whether the defendant should actually be liable, and if so, by how much (as to the entire class).

For defendants, the more important of these two stages is the first because even a nonmeritorious plaintiff can be an enormous risk once a class action has been approved by the court. If the class is large, so will be total damages. Accordingly, even if they are innocent (or believe themselves to be innocent), defendants in class actions where a class has been certified may simply choose to surrender and settle, in effect paying a ransom to escape the perils now threatened by the lawsuit. This, too, is evidently an incentive for enterprising litigators to seek out such cases.

One possible way to reduce the incentives for enterprising class actions that, in effect, blackmail defendants with deep pockets (on their own or with the backstop of their liability insurers) would be to permit the defendant in such cases a limited opportunity to introduce some evidence of its innocence (if it has any) before the class designation is approved or rejected by the court. The stronger this evidence, the higher the hurdle for plaintiffs if they are to sustain their class action. We encourage the legal profession and economists interested in this problem to think of other ways to address it.^[17]

[edit] Competition and Innovation: Keeping Winners on Their Toes

We come now to the third element of the program we have recommended for encouraging productive entrepreneurship. Unlike the first and second elements, which focused largely if not exclusively on individual entrepreneurs and their inventor partners, the third element deals with large firms and measures to stimulate them to continue innovating. This is appropriate and necessary in “entrepreneurial” economies because large firms play an important and complementary role in such economies, by providing the cumulative improvement of entrepreneurial inventions that enables and encourages the mass of consumers to want and use them. Furthermore, it is only vigorous competition among these large firms that forces them to strive without letup to introduce such improvements before competitors beat them to the punch, resulting in a steady stream of innovation and technical progress.

We will focus on several policy instruments to help assure that this outcome— ongoing innovation, primarily by large firms—continues, beginning with a discussion of the proper role of antitrust law and the importance of continued openness to foreign goods, services, ideas, and capital in applying competitive pressure that fuels the innovation race. One part of the corporate income tax code—that dealing with stock options—also could be improved to prod large firms, in particular, to continue innovating. Ideas developed by university faculties, which already are important sources of innovation but are likely to be even more important in the high technology arena in the future, can be commercialized more rapidly, contributing to entrepreneurial success. All of these suggestions should spur both innovation and commercialization by larger firms. Finally, there is a need to ensure that the United States has a sufficiently well-trained workforce, well suited for the increasingly technical demands of the twenty-first century, to staff innovative large and small firms alike.

[edit] Antimonopoly Rules and Cooperative Innovative Activity

In chapter 5, we discussed the role of antitrust laws, not just in preserving but also in encouraging competition. It goes without saying that these laws should continue to be enforced. At the same time, it is equally important that antitrust authorities give more weight in their enforcement efforts to the benefits of certain sorts of cooperation among firms that serve the public interest. Innovation is an arena in which this is particularly true. Innovation helps to speed elimination of the obsolete. It enables firms to share the heavy costs of the research and development process and makes it easier for the individual enterprise to bear its contribution. Innovation reduces the risks of investment in R&D and reduces duplicative efforts. It also facilitates compatibility of design by making it easier for one firm’s technological advances to work smoothly with those of another while enabling firms to learn from one another in undertaking their own innovation activities.

In practice, antitrust authorities can recognize these benefits of cooperation by applying a “relaxed rule of reason” test when weighing the competitive (or possibly anticompetitive) effects of research joint ventures, research consortia, and even mergers in high-technology industries—all of which may be socially optimal responses to market failures that beset the production and dissemination of knowledge. Indeed, mergers of firms with substantial current market shares in high-technology industries (where fixed costs of research, development, and deployment of new technologies are high) can be presumed to be conducive to long-run efficiency, more so than in industries that are less technology-driven.

In fact, both of the U.S. antitrust enforcement agencies—the Department of Justice and the Federal Trade Commission—have recognized the virtues of cooperation through their intellectual property guidelines.^[18] A more formal announcement of a relaxed rule of reason would also clear away some remaining legal uncertainties.^[19]

[edit] Open Borders to Enhance Competitive Pressure

Openness to trade and foreign direct investment is an economic arrangement that arouses the suspicion, and sometimes the enmity, of the general public. It is widely feared that putting out the welcome mat for other countries’ exports, particularly countries with low wages and living standards, will lead to the loss of domestic jobs, loss of business, and depression of wage levels toward the pitiful levels of developing countries. For many decades, popular opinion makers have railed against the unfair competitive threat posed by importing goods made by “cheap foreign labor.” Now, a similar complaint has been lodged against “outsourcing” (or, more accurately, “off-shoring”) of work to service providers in countries paying lower wages.

There are wrong and right responses in rich countries where work is being off-shored. Shutting the door on trade—even partly, via tariffs, quotas, or other impediments to imports—is one rather sure wrong response. The standard conclusion of economists that there are generally mutual gains from trade is fundamentally correct (although, as Baumol and Gomory [2005] have shown, foreign competition can reduce the share of those gains received by countries that off-shore jobs without offsetting advances in innovation). Trade gives importing countries products that manufacturers in other lands can produce more economically in exchange for items made by less costly producers in the exporting countries. Denying the benefits of this beneficial exchange to ourselves—a process that we welcome and openly encourage when all parties to the exchange are located inside the United States—can only add to whatever pains open trade and off-shoring already would inflict upon us.

The right response is for producers in countries where firms are engaged in off-shoring (such as the United States) to innovate more rapidly, developing ever-better and cheaper products. There is emerging evidence that companies that outsource or off-shore some or much of their routine work are exploiting their comparative advantage in bringing innovations to market more rapidly and more cost-effectively (Engardio, 2006). As innovation advances, nations can increase trade and at the same time experience the benefits of economic growth, the creation of more jobs, and the rising overall wages. This outcome is evident from recent United States history. As a share of total output, imports have risen from just 11 percent in 1995 to nearly 17 percent in 2004. Yet during this period, median family income increased by 8 percent, GDP rose by more than 30 percent, and total employment by 12 percent.^[20] Clearly, nations can increase trade and enjoy improvements in their standard of living at the same time.

Indeed, freer trade, coupled with direct investment by foreigners who build their own plants, keeps U.S. firms on their competitive toes and enables U.S. consumers to reap the benefits of a broader range of products and services, available at lower prices. For example, without the competition from more fuel-efficient and higher-quality Japanese automobiles, U.S. auto companies would be producing less fuel-efficient and less reliable autos than they are today. The same process has occurred in the steel and textile industries, among others.

This is not to deny the legitimate concerns about foreign progress in technological fields. There continue to be “first mover” advantages, which can make it difficult (but not impossible) to dislodge firms that are the first to achieve commercial success in new fields. Economies of large-scale production, the lessons learned by an experienced labor force, and the mutual support that related industries derive from one another all can give the early incumbent industries and the nations that host them a powerful defensive position against incursion of technology from other countries. Location still offers advantages even in the age of the Internet, which enables numerous products to be designed and assembled from all parts of the world. Once a locality or region obtains a firm place in some specialty, it seems to embark on a virtuous cycle: the pool of specialized labor and related support services attracts more firms, and this in turn attracts more such people and infrastructure and so forth. Examples are found not only in the United States (most prominently in Austin, Boston, Raleigh, Seattle, and the San Francisco Bay area), but also around the world: Bangalore, India (for software), northern Italy (for design), and Taiwan (for manufacture of consumer electronics). Richard Florida of Carnegie-Mellon University has pointed to this geographical agglomeration as evidence that the world is “spiky” rather than “flat,” as argued by New York Times columnist Thomas Friedman (Florida, 2005; and Friedman, 2005).^[21]

At the same time, it is important to put the concerns about the diffusion of economic progress throughout the world in some perspective. As Americans, we should want the Indias and the Chinas of the world to grow because the richer they are, the larger the markets they provide for our exports. And, just because new ideas are developed elsewhere does not mean that Americans derive no benefits from them. We can gain, just as the rest of the world has been benefiting from our technological progress for decades, by importing capital goods that embody technical change and by accepting foreign direct investment that typically comes with cutting-edge process and/or managerial skills. In fact, the United States has benefited from precisely this sequence of events throughout its history—from British railroads in the nineteenth century to Japanese “just in time” manufacturing and quality circles more recently.

In the long run, all citizens of the world, including those in rich countries like the United States, can expect to be better off if there are more skilled people working to advance the technological frontier. As the popular author Malcolm Gladwell recently put it: “With the pace of development in China and India and other parts of the developing world, we’re just adding to the available brainpower and unlocking these large populations of people and their ingenuity and giving them an education. How much easier will it be to solve the problems of the world when we’ve got 10 times as many brains working on them” (Time, October 24, 2005, p. 86; emphasis added).

All this is not to sound defeatist, but to caution against hysteria. Of course, U.S. leaders in both the private and public sectors should strive to maintain U.S. leadership in innovation and commercialization. We will be better off with this scenario since we will benefit both from the profits generated by innovation and from the virtuous circle of being an economic leader. But closing off our borders to foreign goods, services, ideas, and capital cannot help us in any respect. On the contrary, by depriving U.S. consumers and U.S. producers of what is best in the world, a protectionist retreat is the surest route to second-rate economic status.

The key challenge for policy makers in the United States (and other rich countries) is to maintain a commitment to freer trade in the face of mounting concern across the political spectrum and among both blue-collar and white-collar workers about the threat that foreign competition poses to their job security and earnings advancement. This challenge would be easier to meet if the United States trade deficit were not so enormous. But this deficit is not likely to come down appreciably without major reductions in the long-term federal budget deficit, which acts as a drain on national savings and thereby requires foreign capital to finance United States investment. The counterpart of capital inflows is the high and rising current account deficit. In the meantime, whatever action or inaction may take place on the budget front, federal policy makers could reduce workers’ anxiety about global competition if the government did more to cushion the loss in earnings that workers suffer upon being displaced from a job. One way to do this is to provide limited wage insurance that compensates permanently displaced workers for some portion of any wage losses suffered upon accepting a new but lower-paying job. The United States Congress adopted a limited experiment along these lines in extending the president’s trade negotiating authority in 2002, but only for workers over the age of fifty who could prove that trade was the dominant reason for their displacement. This program should be expanded for all workers, regardless of the reason for their displacement (which is most often the result of continued productivity improvements) and could be implemented at reasonable cost, perhaps less than $5 billion per year (easily financed by a modest increase in the federal unemployment insurance tax).^[22] As we discussed in chapter 7, a similar program should be adopted in Europe as a way to effect a partial transition away from highly costly unemployment insurance programs.

[edit] Government Policies to Facilitate Transfer of Foreign Technology

The United States should remain open to foreign goods and ideas, and its government can and should take an active role in facilitating the use of foreign technology by U.S.-based firms. In fact, no country is an island in the global economy. No country has a monopoly on all innovation. Even the United States and Japan, by far the world’s leaders in the number of patents awarded every year, each create no more than some 35 percent of the world’s innovations, meaning that they must acquire (by purchase or by license) a substantial proportion of the remaining 65 or so percent in order to avoid falling hopelessly behind their competitors.^[23] In fact, technology transfer is inherently an entrepreneurial activity. It entails recognition of foreign technological advances that are promising for importation, redesign of the innovations to adapt them to domestic needs, and then rearranging them for their introduction into the importing country’s productive ventures.

Government can and should play a very helpful role in all of these activities. The main reason for this lies in the fact that new information can be very costly, difficult, and time-consuming to acquire, involving the hard work of many highly educated and skilled individuals. But once these initial expenses are paid, delivery of the information so acquired to others can be virtually costless and the same whether to ten, a thousand, or even more users. In other words, the cost of providing the information to an additional user is virtually zero. Because of this, a central gatherer of information— such as a governmental body—can simultaneously serve the needs of a multiplicity of users at costs far lower than those that would be incurred if each user were to seek to obtain the information for itself. For example, the work of monitoring foreign technical journals and of providing translations of pertinent technical articles can be carried out nearly as cheaply for a multiplicity of that economy’s firms, or even for a considerable number of industries, as it can on behalf of any single business enterprise.

Countries appear to differ substantially in the quantity of resources they devote to this purpose, and the United States is clearly not a leader. Although the following survey is more than fifteen years old, we doubt that the main thrust of the results is any different today than when the survey was conducted. Edwin Mansfield reported in 1990 on the basis of a survey of one hundred American firms in thirteen industries, that these respondents believed only 29 percent of U.S. firms spend as much (as a percentage of sales) on the monitoring of foreign technology development as their Japanese rivals. Forty-seven percent of these American firms reported that they spent as much as their German counterparts, 51 percent as much as their French rivals, and 70 percent reported spending as much as their British rivals on the monitoring of foreign technology (Mansfield, 1990, 343). Such disparities, assuming they continue to persist today (and we suspect they do), may well constitute an opportunity for a growing economy to gain a differential advantage in its monitoring and adoption of foreign technology.

One way to narrow the technology gap would be for the federal government to establish an office of technology monitoring, with a staff of specialists qualified to monitor, translate, and disseminate pertinent materials from foreign publications. Such an office could be lodged in an existing agency, such as the Commerce Department, or even be an arm of the Patent and Trademark Office (provided the new office was given additional funds and existing overworked PTO personnel were not diverted from their current tasks). The office also need not be large. Much of its work could perhaps be outsourced to U.S. universities. In addition, the federal office could be aided by a small army of technical personnel lodged in U.S. embassies around the world, whose main job would be to monitor technological developments—in technical journals, patent applications, and company newsletters and reports—and to report back regularly to Washington. The results could be compiled in a database, translated into English, and made widely accessible through the Internet. The embassy specialists could also help facilitate technology transfer agreements between U.S. and foreign firms and native enterprises.

Universities represent another potentially important transmission link for transferring foreign technology to the United States. For much of the post–World War II period, foreign students traveled to the United States to study at our best universities, often at their government’s expense, gaining cutting-edge technical and managerial skills and knowledge to be put to use in their home countries. Indeed, the balance between native and foreign-born individuals gaining science and engineering Ph.D. degrees from U.S. universities has shifted markedly over time: in 1966, 77 percent of such degrees were awarded to native-born Americans, but by 2000, this figure had fallen to 61 percent (Freeman et al., 2004). Since the terrorist attacks of September 11, 2001, U.S. immigration authorities have made it far more difficult for students from abroad to continue their science and engineering studies here. We believe that this is a mistake and that U.S. immigration policy should be more accommodative to foreign students, granting long-term work visas, or even immediate citizenship, to foreign students who earn degrees in math- and science-related fields (Becker, 2005; and Schramm and Litan, 2005). In 2006, U.S. policy makers agreed with this line of argument. Congress lifted the annual cap on “H1-B” guest worker visas for immigrants in science and engineering from 65,000 to 115,000, and during the summer of 2006, it was debating whether to let more foreign students become permanent residents (we would prefer automatic citizenship) after obtaining advanced degrees in math, engineering, technology, and the physical sciences in the United States.

At the same time, we also urge the U.S. government to learn a lesson from other governments by providing more scholarship aid for American students to study engineering and science in foreign universities. Such a program would extend the existing Fulbright postgraduate fellowships and reach down into the undergraduate and graduate schools, and it would provide subsidies for U.S. students to learn the languages of the countries where they intend to study. In the increasingly global economy of the twenty-first century, it is only appropriate that formal government policy enable U.S. students to be better prepared for working and starting businesses that compete globally. Recognizing that we have much to learn from the rest of the world is a good way to start.

Finally, technology transfer could be accelerated not only by sending Americans abroad, but also by welcoming highly trained foreigners here. Such a policy may be stoutly resisted by their high-skilled United States counterparts, whose earnings growth could be diminished by the added supply of well-trained workers—just as lesser-skilled workers typically object to labor-intensive imported goods or the immigration of low-skill workers themselves. But in a world where knowledge increasingly has no limits, it would be a collective mistake not to take advantage of the training and skills of workers from abroad.

[edit] Encouraging Innovation through the Corporate Income Tax System

Like the individual income tax, corporate income taxes can be used to encourage entrepreneurial firms. We have in mind one modification to corporate income taxes—whether or not those taxes are increased, reduced (as many economists have long argued), or left roughly where they are now in terms of raising revenue—that should help catalyze innovation both by large companies and among newer, more entrepreneurial firms.

Our specific subject is stock options, which since the early 1990s have become an increasingly popular means of compensating individuals, especially at high-growth, high-technology companies. Stock options afford their recipients the opportunity to buy a company’s stock at a given price (the “strike price”) and thus provide a potentially significant upside opportunity if stock prices increase but no downside risk (since the holders of options have no incentive to exercise them if stock prices do not increase). In principle, stock options help address what has come to be known as the principal-agent problem, with the stockholders being the principals and corporate managers their agents. The “problem” in this arrangement arises from the possibility that the agents will look after their own interests in the company and attend only secondarily, if at all, to the interests of the company’s owners. By giving corporate managers an ownership stake, or at least a potential ownership position (once options are exercised), stock options theoretically better align the interests of corporate managers and shareholders.

But there is an important drawback to stock options, as they currently exist. If the value of the options is tied only to a company’s stock price, and the stock price in turn can be temporarily boosted by higher-thanexpected earnings, managers may be tempted to “cook the books”—that is, falsify their accounting records by understating expenses or overstating revenues, in an effort to beat earnings expectations (particularly the expectations of the stock analysts who follow the fortunes of public companies) in the hope of cashing in their options after the stock price has risen. The corporate scandals of the late 1990s and early 2000s can be explained by just such a sequence of events (Bogle, 2005; Bromwich et al., 2002).

Some of the abuses of stock options are likely to be corrected by the 2005 decision of the Financial Accounting Standards Board (FASB) to require public companies in the United States to begin recording stock options as a compensation expense, using any one of several approved techniques. The FASB took this step after a more than a decade of debate over the issue and after earlier congressional pressure against such a measure, largely at the behest of high-technology companies (which feared that “expensing” of options would inhibit them from attracting talented employees and executives). There is insufficient evidence to know whether this fear is or ever was justified, but even if it were, it could be addressed by exempting new public companies from the expensing requirement for a limited time after their listing on a public exchange.

The issue we want to address here, however, is how stock options are or should be treated for tax purposes. Unlike the FASB provisions, which essentially require companies to estimate expenses associated with stock option grants, current corporate income tax provisions allow deductions only when employees actually exercise the options. The value of stock options for tax purposes differs from their value for financial reporting purposes most likely because it is easier to verify stock-option expenses based on the difference between the exercise and strike price (the tax definition) than some estimate of the value of the options when they are granted (the reporting definition). More important for our purposes is that U.S. tax law permits companies to deduct expenses associated with stock options regardless of how the firm itself has performed relative to other firms in the same industry or with respect to all other publicly traded firms. Stock options structured in this manner (as virtually all options are) do not reward managers and employees for extraordinary performance; on the contrary, option holders can benefit from a sub-par performance by their companies since they can realize gains from the options as long as the price of their company’s stock rises.

Furthermore, although companies generally restrict managers and employees from exercising their options for a period of time after they are granted, this limitation is often as short as six months. As a result, option holders have an incentive to maximize the short-run performance of their company’s stock (including “cooking the books”) regardless of the effect their activities may have on company performance in the long run.

We believe that the corporate tax law would better encourage extraordinary, long-term company performance—and thus innovation—if stock option expenses were deductible only if they meet the following conditions: (1) stock options cannot exercised for some substantial period, say five years, so that anyone who receives them will think beyond the shortterm effects on stock prices; and (2) stock options must be performance based, with their amounts contingent on the firm’s exceeding the performance of comparable firms or its own past record.^[24] Separately, the Securities and Exchange Commission, which governs reporting by public companies, should require the prompt disclosure of stock sales by top management so that the investing public knows when “insiders” are selling their shares.

The proposed tax reform coupled with the proposed disclosure requirement should discourage managers from taking steps that may artificially boost a company’s stock price and then taking their chips off the table by selling their shares. Viewed more positively, the proposed reforms should make clear to corporate managers that the only way they can enhance their own compensation is to improve their companies’ pace of innovation, which is good for shareholders and for the economy as a whole.

[edit] Encouraging the Commercialization of University-Based Research

Entrepreneurs clearly must have incentives to bring innovations to the marketplace. So must large companies, although, as we have argued, their innovations are more likely to be incremental in nature and forthcoming only so long as these firms face competitive pressure to continue the innovations arms race.

Increasingly, at least in the United States, universities also have become sources of new technological knowledge and new products and processes.^[25] As technology becomes more complex, universities may play an even more important role in advancing innovation by generating ideas that fuel the innovations arms race among large and small companies alike. Until 1980, there was some uncertainty about the extent to which universities and their faculties had rights to their innovations, which many believed discouraged their commercialization. This uncertainty was cleared up with the passage of the Bayh-Dole Act, which explicitly allowed universities and their faculties to commercialize, and thus to profit from, federally financed research. The aim of the legislation was to move university-based science “out the door” and into the commercial marketplace as rapidly as possible, whether through the formation of new ventures or the sale or licensing of the technology to large firms, many of which look to universities to fill their own innovation pipelines.

By a number of indicators, the Bayh-Dole Act appears on the surface to have been a success, gauged by a fourfold jump in the share of all domestic patents accounted for by universities from 1980 to the mid-1990s (Mowrey, 2005),^[26] and the visibility of some of the high-technology companies that have spun out of universities: Google (Internet search), Genentech (biotechnology), Chiron (vaccines), Cirrus Logic (information technology), and Netscape (which developed the first widely used Internet browser, only to be eclipsed by Microsoft, in part through conduct that the United States courts later deemed to be in violation of the U.S. antitrust laws). Large companies and venture capital firms also eagerly line up outside university doors, not waiting to acquire the spun-out companies, but rather seeking to license the universities’ best technologies from the outset. In recent years, other countries (which heretofore have discouraged the commercialization of university-developed innovations on the grounds that this would distract them from their core functions of teaching and basic research) have been seeking to implement or have already adopted their own versions of the Bayh-Dole Act.

For all this seeming success, it is our view that university research is not flowing as fast as it should to entrepreneurs and to larger companies that are trying to become more entrepreneurial by embracing and exploiting newly developed technologies. The lure of the rewards from commercialization seems to motivate only a handful of researchers at a relatively small number of universities. For example, only four universities generate more than 100 patents every year, and only ten report more than 270 disclosures, which represent the pre-patent stage of a new discovery. A summary of where the commercialization action really is taking place, showing how limited it is, is shown in table 18.

We believe that universities—even those where innovation and commercialization activity is already concentrated—could do a much better job of moving their discoveries to market if they were less bureaucratic, and more entrepreneurial, in doing so. One of the unintentional by-products of the Bayh-Dole Act is that universities somehow have come to believe that they can only, or at least best, implement the act by centralizing their commercialization activities in their in-house technology transfer offices. Universities had good reason for believing this. In principle, such offices should be able to help enterprising faculty members determine whether their innovations are patentable or protected by some form of intellectual property right and then pay for the necessary legal work to file the patent applications and negotiate licenses or uses in commercial startups. Accordingly, university technology trade offices, in theory, should be able to accelerate the commercialization of university-based innovation.

But research work documented by staff at the Ewing Marion Kauffman Foundation has revealed that in practice (with few exceptions) many university technology trade offices have become bottlenecks to commercialization. Given their situation within university bureaucracies, technology trade offices have difficulty paying for and thus attracting the sophisticated negotiating, licensing, and legal talent that otherwise moves to the private sector. In addition, the typical university technology trade office is understaffed, which inevitably slows the commercialization process. As a result, although they are technically bound to clear their commercialization activity through the technology transfer office, many university faculty members commercialize through the “back door,” circumventing that office altogether (Shane, 2004; Audretsch et al., 2006). Furthermore, it is rare for a technology transfer office to generate sufficient income to cover its expenses, again with certain exceptions for the universities with substantial licensing income.

In our view, the commercialization process could be much improved if universities experimented with a number of different arrangements. To reduce the delays and transactions costs associated with licensing negotiations, universities could jointly develop or follow the lead of others to develop standardized licensing agreements. An obvious analogue is the typical publishing contract that contains a standard royalty rate applied to sales through different media. The royalty rate, and possibly other terms, might vary depending on whether the license is exclusive or nonexclusive. Admittedly, it may not be possible to standardize start-up arrangements, which most likely will continue to require custom agreements. But this process, too, could be accelerated if the technology transfer office were subjected to some competition.

One of the reasons technology transfer offices fail to generate a profit for their universities, which would seem to be a principal rationale for their existence, is that they can neither reach a size sufficient to realize economies of scale nor attract licensing and other personnel with expertise comparable to that available in the private sector. These problems might be addressed if universities in a region (at the very least public universities) combined their resources to share a single technology transfer office, rather than each continue to maintain its own office. To be sure, there would be up-front and perhaps continuing problems of “turf ” in a sharing arrangement, but assuming these could be mostly resolved in advance (via agreements about which institutions get “credit” for which inventions), the economies of scale in such joint arrangements should clearly offset any operational difficulties.

Another notion is for the universities to change the objectives of their technology transfer offices—from maximizing profits (or at least minimizing losses) to maximizing the volume or numbers of “deals” that are arranged. Knowing that they will be graded by how many innovations their offices “get out the door” might loosen the bottleneck and reduce the delays that are now created when the transfer offices concentrate so much of their scarce time and money on negotiating the “perfect” licensing arrangement for one or a few technologies while ignoring the innovations that pile up, waiting for attention.

A fourth, more radical idea would be for universities to experiment with ending what amounts to the technology transfer office monopoly, which stems from the requirement that all faculty members use that office for any licensing or other commercialization of their innovations. Instead, universities could let faculty members choose between their own intellectual property agents or the university’s technology trade office. Allowing competition should spur the transfer office to improve its performance, reducing delays and transactions costs.^[27]

Perhaps the most radical suggestion of all is that universities, in some fields, simply abandon their quest for up-front profits and make certain intellectual property freely available, along the lines of the open-source model developed so successfully by Linux for computer operating system software. Such a policy would accelerate the introduction of new ideas into the marketplace. To the extent universities harbor any desire for future profits from the successful commercialization of ideas generated by their faculty, they could hope to realize them in a different form later, perhaps through generous gifts by individuals or companies who profit from them. The idea is not far-fetched. As it is now, universities rely heavily on grateful alumni for significant sources of endowment or operating expenses. Universities can realistically expect similar contributions from grateful faculty who have full rights to commercialize their innovations.

Though seemingly disparate, these suggestions have a common objective: to speed up the introduction of university-generated ideas into the marketplace. This can benefit the universities financially, but it is also an important social good, especially when, increasingly, companies headquartered around the world are deciding where to locate their R&D facilities, based in part on their ease of access to university-based research personnel and output (Thursby and Thursby, 2006).

Why, then, have university presidents not already adopted one or all of these measures? We suspect the answer, in part, is that most presidents are not themselves familiar with the commercialization process and treat it much like information-technology decisions, delegating them to some subset of individuals presumed to be knowledgeable in the area. In the case of IT, this may be a sensible strategy. But when it comes to the commercialization of technology, the danger of delegation to an independent office is that the manager and employees of the delegated office, namely, the technology trade office, may acquire vested interests of their own, seeking to preserve their own jobs and possibly expand their operations. Given the many responsibilities of university presidents, it is not unexpected that they would largely ignore what the technology trade office is doing and treat any commercialization as a success—as a “free good,” one might say—without realizing what could be realized if the commercialization process were organized differently.

Here is where the federal government could play a constructive role. At a minimum, the key agencies that award significant research funds to universities (the National Institutes of Health, the National Science Foundation, the Department of Energy, and the Department of Defense) could collectively agree (perhaps with the guidance of the White House science advisor and/or the director of the National Economic Council) to convene university presidents to inform and persuade them of alternative models of commercialization. A more aggressive approach would be for the same agencies to condition their research monies on universities experimenting with one or more approaches to accelerating commercialization. However it is accomplished, when more universities become skilled at transferring technology, and when they become more efficient and inventive as organizations, entrepreneurs and large firms, the United States, and the rest of the world will benefit.

[edit] Maintaining a Well-Trained Workforce

The living standards enjoyed by residents of any economy depend on the productivity of its workforce. Workers are more productive if they work with more and better capital equipment. They are also more productive if they have the benefit of education and training.

Education and training also play a key role in the innovation arms race that is essential to sustain economic growth in any economy. Firms that have succeeded in “round one” cannot be expected to continue their success in subsequent “rounds” of competition unless their managers and workers have the skills necessary to generate innovations or, at the very least, to recognize and purchase the rights to innovations developed by others.

With the coming of the twenty-first century, Americans have grown worried about the quality of their elementary and secondary educational systems and their ability to continue to provide new generations of Americans with the requisite skills. A number of commissions, chaired by highly qualified and influential individuals, have pointed to a series of disappointing statistics about student achievement in mathematics and science, in particular. According to various international tests, American students rank well behind those in other parts of the world in these subjects, although there continues to be some debate about how poorly American students do on average, and whether their comparative performance declines in upper grades or essentially remains relatively flat as they age.

Meanwhile, at the university level, there seems to be declining interest in mathematics and science among our native-born youth, and the numbers of foreigners who in recent years have made up a substantial share of our mathematics and science graduates are now down as a result of immigration restrictions in the aftermath of the terrorist attacks of September 11, 2001. The United States was once first in terms of the percentage of its population of twenty-five to thirty year olds that attained tertiary education, but by 1991 it had fallen to third place among OECD countries, and by 2003 it had dropped to eighth position (Bowen, 2005).^[28] All in all, one can paint a disturbing picture of the status of American education at all levels, with other nations catching up to us in educational attainment, paying much lower wages, while our system for training our future workforce seems plagued with problems. This situation seems especially dire given the increasing tendency for U.S.-educated foreign scientists to return to their home countries, where state-of-the-art research facilities often await them (Lemonick, 2006).

But the source and consequence of the challenges the United States faces must be properly understood. The K-12 statistics, as alarming as they are, are average rankings, mixing our best and brightest in excellent suburban public and private schools with many urban schools that are performing poorly. The real worry about the apparently poor math/science precollege performance of U.S. students may be more about what it portends for future income inequality than what it means for continued technical change.

Meanwhile, the concern about the apparent inadequacy of engineers in the United States misses several important facts. For one thing, although the United States trails China and India in the absolute number of engineers produced at the college and graduate school levels, on a per capita basis, the United States still ranks well above both countries.^[29] This should continue to be true for some time, if not the indefinite future. Second, the comparisons of absolute numbers do not take account of the difference in the quality of engineers produced in the United States and other countries. Although China, India and other countries have first-rank universities producing their engineers, the United States has many more such places of higher learning, turning out very well-trained engineers and scientists. And even if the rest of the world is producing and using more scientific talent, in the long run, this will benefit both other countries and the United States, for reasons we gave earlier. Finally, the marketplace in the United States is not signaling a shortage of engineers, for otherwise firms in need of their services would increase their salaries more rapidly than they have in the past, thereby inducing more people to enter the field. Policy makers would do well to heed what happened the last time alarm bells seemingly went off about the shortage of engineers: in 1987, when the National Science Foundation predicted a shortage of engineers, it prompted Congress to increase scholarship funding and to expand foreign visa allowances, resulting in a glut of American-trained engineers several years later, many of whom had difficulty finding jobs (Friel, 2006, 40).

Still it is likely that, as a group, engineers and scientists generate social benefits that exceed the private benefits that accrue to them; in this respect, engineers and scientists are like teachers. This is one reason why it makes sense, in our view, for government to continue to subsidize their education, as well as to provide research funds through universities, which indirectly increases the demand for their services and thus their job prospects and their earnings. As it is, federal support for basic R&D has fallen steadily over the years, from 2 percent of GDP in 1965 to less than half that in 2005. Further, the composition of federally supported R&D has shifted over time toward health sciences and away from the physical sciences, engineering, math, and computer science,^[30] a trend that the Bush administration’s proposed federal budget for fiscal year 2007 would begin to reverse over the subsequent ten years. Industry R&D spending has fared better than federally supported R&D, but this is for applied research rather than the basic R&D that is more likely to spawn radical breakthroughs (Friel, 2006).

Whether the United States has too few, too many, or the just right number of scientists and engineers, there is little question that U.S. firms will demand better math and science skills of their other workers over time. Those without this training will find themselves at a competitive disadvantage in the workplace and thus cannot look forward to enjoying the living standards of those who have these skills. As long as skills are highly unevenly distributed across the population, so will be incomes. The standard prescription for rectifying these educational imbalances, of course, is to spend more money. Several prominent reports, as well as President George W. Bush himself, have offered various proposals for increasing funding— for more math and science teachers, equipment, remedial training, and so forth.^[31] We are sympathetic with this approach only to the extent that any additional educational funds are channeled to teachers under the kind of performance-based pay systems that are prevalent in the private sector. Otherwise, additional funds have little prospect of improving educational performance among any set of students.

More fundamentally, as much if not more attention needs to be paid to how education money is spent than to the total sums involved. As contentious as educational policy is, it is generally agreed that America is not getting the largest “bang” out of its hundreds of billions of education “bucks” as it could. It is appalling, for example, that there is so little research on what educational practices actually work to improve the performance of underprivileged students. Our educational practices are much like health care was before the nineteenth century, when doctors proceeded without evidence and resorted to little more than bloodletting and cupping as the universal remedies for most illnesses. Whatever else is done, this lack of research on educational best practices must be remedied. Furthermore, improving educational performance for all students, but especially those from disadvantaged backgrounds, requires the promotion of more competition at the local level. Charter schools are a form of competition, but we would prefer to see at least some form of the voucher system recommended long ago by economist Milton Friedman. If vouchers for private schools prove to be constitutionally unacceptable, then, at a minimum, parents should have choices within the public schools about where to send their children.^[32] And if sufficient choices are not available, then, again at the very least, public school authorities should treat public charter schools on equal footing with other public schools, by providing them with the equivalent amount of per pupil funding as is now directed to conventional public schools.

[edit] The Political Economy of Growth

One problem with many books like this one is that the policy suggestions they provide risk looking impractical or politically infeasible, at least in the short run. It is a sad but unfortunate fact that in any democracy, political leaders rarely anticipate problems but instead react to crises, real or manufactured. The critical question is: what policy levers are pulled when those crises occur? The answer, more often than not, is that policy prescriptions or ideas that have been percolating in the academy or in the think tank community are taken off the shelf and dusted, as it were, and written into law in some form. Examples of crises-led policy change abound throughout American history: the securities registration and disclosure laws of 1933 and 1934 and the Glass-Steagall Act of 1933 that walled off commercial banking from investment banking were adopted only after the Great Depression was already underway; the banking reform laws in the 1980s and 1990s were enacted after the wave of failures of depository institutions in the 1980s; and the Sarbanes-Oxley Act was adopted only after various corporate scandals in the early 2000s. Such steps typically close the barn door after the horses have escaped and in some cases turn out to be counterproductive yet difficult to repeal or modify (the best-known example being the Glass-Steagall Act, which Senator Carter Glass, one of its sponsors, urged be repealed only two years after enactment, advice that was not taken for more than sixty years).

Policy also changes when political leaders successfully “manufacture” crises over long-festering problems in order to create the urgency needed to mobilize policy action, especially when congressional action is required. Examples here include the War on Poverty in the 1960s; the tax code simplification of 1986; and welfare reform in the 1990s, which represented a partial undoing of the 1960