CHAPTER 4

THINGS GET MORE COMPLICATED

In the previous chapter, we highlighted the importance of innovation to economic growth, and the role that young firms, particularly those backed by venture investors, play in stimulating new ideas. When reviewing these propositions, we were able to point to a substantial body of academic research—and many less systematic (but perhaps more persuasive!) stories—suggesting that a strong relationship exists between new ventures and innovation.

But as we noted in the previous chapter’s conclusion, these two facts alone are insufficient for us to come to a definitive conclusion. In particular, if government intervention in this arena is to make sense, it is necessary for the effort to be effective in stirring entrepreneurial and venture activity. And it is by no means certain that this is the case.

In this chapter, we’ll review previous research and what it tells us about the effectiveness of government intervention. Here, alas, the evidence is less persuasive. On the one hand, there are many reasons why governments should be able to play a catalytic role. By its very nature, entrepreneurship is an activity that feeds on itself, which means that a public “jump start” may well be helpful. But on the other hand, government programs to boost venture activity have frequently fallen prey to incompetent decision-making or else to outright distortions by special interests. We’ll discuss these challenging threats to effective government programs in this chapter.

WHY GOVERNMENTS CAN ENCOURAGE VENTURE ACTIVITY

Even if entrepreneurship and venture capital can play an important role in spurring innovation, it is natural to ask why government should intervene in these markets. Isn’t this economic activity one that is best left to the private marketplace? Aside from the historical anecdotes alluded to in chapter 2, what case can be made for government intervention?

Beginning a Virtuous Cycle

The first rationale for government intervention lies in the fact that there is a “virtuous cycle” in entrepreneurship and venture capital. Activities by pioneering entrepreneurs and venture capitalists pave the way for subsequent generations: in a given city, it is far easier to recruit the staff for the one-hundredth start-up, or to find a lawyer to structure the one-hundredth financing, than the first.

Indeed, history is full of examples of pioneering firms that served as “entrepreneurship academies,” from which other entrepreneurs sprung. The most famous example has already been briefly mentioned, Fairchild Semiconductor. The firm’s origins had been Shockley Semiconductor Laboratory, which William Shockley founded in 1956.¹ After failing to lure his former colleagues from Bell Labs, he had hired some of the best graduates from U.S. engineering departments. But Shockley’s autocratic style soon alienated many of the new hires. In 1957, eight key engineers—the Traitorous Eight, as they soon became known—left Shockley and formed their own company, Fairchild Semiconductor. While the company was successful in many respects, introducing the first commercially available integrated circuit and becoming one of the major players in Silicon Valley in the 1960s, it also saw frequent defections by engineers who founded their own firms. Among the semiconductor companies founded by its alumni were AMD (Advanced Micro Devices), Computer MicroTechnology, Cirrus Logic, Intel, LSI, and National Semiconductor—the key players in the industry in the decades to come. Other firms have played similar roles elsewhere, such as Minnesota-based device manufacturer Medtronic and the executive search firm Recruit in Japan.²

Several forms of information generation lead to this “virtuous cycle”:

•  Employees at large firms may initially be reluctant to “make the plunge” and join a start-up firm. Notions such as stock options may seem alien and insufficient justification for the lower compensation and higher risk frequently associated with young firms. Over time, the great rewards that equity in young firms can bring, as well as the other benefits of working for a dynamic young firm, are increasingly appreciated.

•  Much of the entrepreneurial process is an art rather than a science. The surest way to appreciate the issues at work in this setting, and to successfully navigate the many shoals that lurk to scuttle the unwary, is to live through a previous venture. With the proliferation of entrepreneurial ventures, there develops a cadre of seasoned, successful entrepreneurs, who are far more effective managers in these settings than their peers.

•  Entrepreneurs become familiar with the trade-offs associated with venture capital financing. Initial disputes about the types of terms and conditions commonplace in venture financing are balanced with an appreciation for the types of gains possible with the involvement of a seasoned financier.

•  Intermediaries such as lawyers and accountants become familiar with the venture process, and can better advise entrepreneurs and financiers alike.

•  Institutional investors gain greater confidence that the sector in which venture capitalists are operating is a viable one, and become more willing to back funds.

•  Venture capitalists more readily find peers with whom they can share transactions. The syndication of transactions is an important form of “judgment sharing,” which allows a group of venture capitalists to make more effective decisions than if each one operated alone.³

An extensive body of economic thought in public finance discusses the circumstances in which it is appropriate for the government to offer subsidies. These works emphasize that subsidies are an appropriate response in the case of activities that generate positive “externalities,” or benefits to others that are not captured by the firm or individual undertaking the activity. Thus, governments often provide subsidies to firms that invest in pollution control equipment or individuals who install solar power. Most of the benefits from their investment to reduce pollution and greenhouse gasses will benefit all of us, not the firm itself. To encourage investments that primarily benefit other firms and all of society, public subsidies are often an appropriate response.

In a similar manner, pioneering entrepreneurs and venture capitalists generate positive externalities that benefit others. It is precisely when such externalities are present that public interventions–whether tax incentives, regulatory shifts, or more direct measures—are justified. These spillovers to other firms are likely to be particularly important in the youngest days of the entrepreneurial sector or the venture capital industry, when pioneering new ventures and investment groups are just getting established. These relationships suggest that government may have an important role in priming the pump for additional entrepreneurial and venture activity during the industry’s inception. Once the industry reaches a critical mass, a process that will take years or even decades, the case for public intervention will wane.

This claim can be supported by the low returns that many pioneering venture funds have garnered. For instance, venture and growth equity funds active in the developing world—which I define here as countries other than Canada, the United States, western Europe, Australia, Japan, and New Zealand—have garnered very disappointing returns over the past two decades. If one takes the simple average of the rates of returns of these funds, they have generated an annual return of 3.8 percent. The weighted average (that is, when we count larger funds more) is even grimmer: -1.5 percent.⁴ These returns are less than those that would have come from holding the safest Treasury bills, much less public stocks.

A natural interpretation of these patterns is that they reflect the difficulties of undertaking successful investments in markets such as China and Russia, where regulatory uncertainties, inexperienced entrepreneurs, and a problematic judiciary make investment challenging. But it appears that if one looked at the returns from the early days of the U.S. venture industry, one would see a similar picture. For instance, the pioneering venture American Research and Development, whose history we discussed in chapter 2, had an annual return of only 14.7 percent despite its “home run” return from Digital Equipment.⁵ There were many dozens of funds during the decades in which ARD was active, most of which are forgotten. Given that the funds that have faded in the mists of time were in all probability far less successful than ARD, the returns of the U.S. venture industry during its early years are likely to have been no better than the emerging market funds in their initial years. This pattern suggests that no matter how promising the returns of entrepreneurial activity ultimately are, in a venture market’s early years, low returns are likely.

Providing Certification

A second rationale for government involvement lies in its ability to provide a stamp of approval. A growing body of empirical research suggests that new firms, especially technology-intensive ones, may receive insufficient capital to fund all value-creating projects. Why cannot entrepreneurs get funded? A frequent source of blame is information asymmetries. The entrepreneur invariably knows more about the central technology than anyone else. But outside investors cannot uncritically accept the claims that entrepreneurs make, since they have so much to gain from getting funding (and are likely to be excessively optimistic in any case). As a result, great ideas may go unfunded.

Indeed, economists have studied the impact of capital constraints—the inability of firms to raise enough money, typically because potential investors lack sufficient information—and documented the breadth of this problem. An inability to obtain external financing limits many forms of business investment.⁶ Investments in research and development are no exception: capital constraints also appear to limit expenditures, at least in smaller firms.

As we discussed in the last chapter, venture capitalists specialize in financing this type of firm. They address their need for information through a variety of mechanisms. Other investors, aware that venture capitalists are astute investors in these settings, should be able to follow their lead, and back companies that they finance: the certification that venture investments provide should unlock the door to much more funding. Thus, expenditures by the government to catalyze venture funds could have an “add-on” effect. Other investors, confident that information deficits had been overcome, could confidently follow the venture investors’ lead.

Taking this argument even further, government investments may also have a certifying effect. Why not just rely on the venture capital industry to provide a stamp of approval? A primary reason is that venture capitalists back only a tiny fraction of the technology-oriented businesses begun each year. In 2000, a record year for venture disbursements, just over 2,200 U.S. companies received venture financing for the first time; in 2007, the number was 1,279.⁷ Yet the Small Business Administration estimates that, in recent years, about one million new businesses have started up annually.⁸ Furthermore, private venture funds have concentrated on a few industries: for instance, in 2000, fully 46 percent of the funding went to Internet-related companies. More generally, 92 percent of the funding went to firms specializing in information technology or health care. By 2008, the spotlight had shifted to renewable energy, among other topics: almost 16 percent of the funds in the second quarter of 2008 went to companies in the “energy and industry” category. Whatever the flavor of the moment, many promising firms in other industries are not attracting venture capitalists’ notice, perhaps reflecting “herding” by venture capitalists into particular areas, a problem that finance theory suggests affects institutional investors.⁹ If government programs can identify and support these neglected firms, they might provide the stamp of approval these high-potential, underfunded firms need to succeed.

But if government officials are going to address such problems as herding, they will need to overcome the many information asymmetries and identify the most promising firms or else choose venture groups that can. Otherwise, their efforts are likely to be counterproductive.¹⁰ Is it reasonable to assume that government officials can overcome problems that private sector financiers cannot? This possibility is not implausible. For instance, specialists at agencies that concentrate on funding health care and defense research may have considerable insight into which biotechnology or advanced materials companies are the most promising from a scientific perspective (though of course, interesting science and engineering does not always translate into a profitable company), while the traditional financial statement analysis undertaken by bankers would be of little value. In general, the certification hypothesis suggests that these signals provided by government awards are likely to be particularly valuable in technology-intensive industries where traditional financial measures fall short.

Creating Knowledge Flows

A third rationale for public entrepreneurship and venturing initiatives is that knowledge spillovers may result.

An extensive literature has documented that innovation is one area where spillovers are commonplace.¹¹ These spillovers take several forms:

•  For instance, a firm may make a substantial investment in a new product only to see a rival capture most of the sales and profits: think about SaeHan Information Systems, which introduced the first portable digital music player in 1998. While the Korean manufacturer solved the key technical problems associated with the device, its ultimate sales were a tiny fraction of Apple’s iPod.

•  In other cases, another firm that develops a related product may get most of the profits. For instance, the bulk of the rewards associated with personal computers since their inception in the early 1980s have not gone to manufacturers such as Hewlett Packard and Lenovo, or to application developers such as Lotus or WordPerfect, but rather to two firms that contribute other essential inputs into these computers, the microprocessors (Intel) and the operating system (Microsoft).

•  Finally, innovations may end up not being very profitable, while very beneficial to society as a whole. One example may be Amazon, which after a decade of operation has not come close to earning back the capital provided by its investors. At the same time, the firm has made books and other merchandise much more available to many who do not live near major bookstores or specialty retailers.

Thus, in many instances, the firms pursuing an innovation get fewer benefits than society as a whole. As a result, left to their own devices, companies will do less research than desirable. But with government subsidies, firms may be encouraged to invest the socially ideal amount of funds in R&D.

One way that economists try to get a handle on the extent of underfunding of research is to compare the return that a company gets from undertaking research and that which society as a whole earns. Such an exercise is easier said than done: measuring the social rate of return is very tricky, and requires quite a few assumptions along the way. After reviewing a wide variety of studies using quite different methodologies, Zvi Griliches estimates that the gap between the private and social rate of return is substantial, probably equal to between 50 percent and 100 percent of the private rate of return.¹² Thus, if a firm earned a 10 percent rate of return on its own investment in research, society would be earning 15 to 20 percent.

While few studies have examined how these gaps vary with firms’ characteristics, a number of case-based analyses suggest that the differences between social and private returns are especially large among small firms.¹³ These organizations may be particularly unlikely to effectively defend their intellectual property positions or to extract most of the profits from their discoveries when competing against larger firms. As a result, it may make sense for governments to fund young research-intensive firms, even if the direct financial returns from these investments are somewhat less than would be reasonable given the risks that are taken on.

Taken together, these arguments suggest that subsidies to entrepreneurs and venture firms can have multiple benefits. But, as we will soon see, the logic is not so straightforward.

THE CASE AGAINST GOVERNMENT INTERVENTION

The arguments outlined above implicitly assume that, once a problem needing public intervention is identified, the government can dispassionately address it. But this is a substantial leap. Government officials with the best of intentions can take counterproductive steps, and in some cases, their intentions are not the best. Distortions may result from government subsidies, as interest groups or politicians direct subsidies to benefit themselves. In this section, we’ll explore how intervention can go very wrong.

Incompetence

These basic insights have been developed in an extensive literature on political economy and public finance. The first concern has to do with the competence of government. In many instance, officials may be manifestly inadequate to the task of managing entrepreneurial or innovative firms.

Much of the literature has addressed the quality of governmental efforts in general, rather than focusing on programs to enhance venture activity. Nonetheless, it suggests several clues to where the ability successfully to design and implement investment initiatives is likely to be found:

•  Nobel laureate Douglass North has argued that as nations become wealthier, their ability to invest in government institutions grows.¹⁴ Moreover, citizens and businesses are likely to demand better governmental services. As a result, nations with more wealth per citizen should have better governments. (Of course, the relationship could actually go the other way: countries could become rich because they have better governments.)

•  Several political scientists and political economists have argued that ethnic homogeneity in societies is associated with better governments.¹⁵ In diverse societies, the political winners all too often have focused their energies on expropriating wealth (or worse) from the other ethnic groups and enact measures that reinforce their hold on power. These steps are unlikely to lead to good government, to say the least!

•  A growing “law and finance” literature suggests that there is a considerable amount of historical accident, or what economists sometimes call “path dependency,” in how governments work.¹⁶ For instance, many nations today adhere to either the common law or civil law tradition, legal systems that originally developed in England on the one hand and France and Germany on the other, and then spread through the world via colonization, conquest, or voluntary imitation. These traditions appear to have powerful impacts on how governments work, though it is not immediately obvious how these differing legal systems play out.

A 1998 paper by four leading economists in the “law and finance” literature tries to assess the extent to which these three considerations shape broad measures of governmental performance.¹⁷ The results indicate a positive correlation between per capita income and government performance: rich nations, not surprisingly, have better governments than poor ones. In addition, more homogeneous countries, measured using languages spoken and ethnicity, have better governments than more diverse ones. In addition, government origin seems to matter: common-law countries, such as Great Britain and the United States, have better governments than civil law (particularly, the French civil law) or socialist law-based countries, highlighting the importance and influence of historical circumstances.

A flagrant example of government incompetence in promoting innovative activities comes indeed from France. The French government’s efforts to encourage high-technology entrepreneurship over the past few decades have seen, to put it kindly, a series of miscues.

Consider, for instance, efforts to promote the electronics industry in the 1980s.¹⁸ Following the ascension of François Mitterrand and the Socialist Party in 1981, the government spent about $6 billion to acquire a number of lumbering electronics giants, including CII Honeywell Bull and Thomson. Meanwhile, a number of promising smaller firms in the industries were either acquired directly by the government or pressured into merging with the giants.

The results were an unmitigated disaster. At the existing firms, once the government subsidies were in place, a tide of red ink turned into a torrent, with annual subsidies for annual losses growing from $226 million in 1980 to $4.6 billion in 1982. The vast majority of the ideas championed by young firms were extinguished as they became part of stultifying bureaucracies. Nor did the government put any real pressure on the established firms to develop their younger partners’ ideas: the public bureaucrats’ single-minded focus was on preserving employment at large existing factories. The contrast with Taiwan’s successful efforts to stimulate its computer industry in the 1990s, where numerous subsidies were given to small firms with the expectation that many would fail but a few succeed brilliantly, could not be more stark.¹⁹

Even if we look just at the primary goal of the French government, their efforts to preserve jobs at existing French computing employers were essentially futile. The government was forced to sell off many firms, with attendant job losses, in the face of a political uproar over the size of the subsidies. Even companies that it continued to hold, such as Bull (in which the government held a majority stake until 1997), employment fell to 8,000 from a peak of 44,000 in 1991.

Nor are such ill-conceived interventions confined to the distant decades in France. Consider the efforts to create a high-technology cluster in Brittany.²⁰ The motivations that drove this effort were reasonable. Brittany had long been a hub for naval shipbuilding, indeed since the seventeenth century. As the French government cut back sharply on military procurement in the mid-1990s (domestic defense acquisition spending fell by nearly 40 percent between 1993 and 1997, even before adjusting for inflation), the dislocation that cutbacks would cause was a natural political issue.

But the solution—to create a French Silicon Valley in Brittany, with a focus on electronics—was remarkably ill-considered. Not only was there a limited entrepreneurial tradition in this region, but the shipyard was a significant share of all industrial activity. Recall that we saw in chapter 2 that both the region’s entrepreneurial tradition and industrial backbone were important in the creation of Silicon Valley. But rather than a nascent hub of high-productivity innovation, Brittany was—and remains—dominated by lower-productivity industries.²¹ This development route appears to have been decided upon in Paris, with little consultation with either local political leaders or potential entrepreneurs or business leaders. Moreover, after announcing an ambitious initiative for Brittany, funding was trimmed as more and more regions were highlighted for incentives.

One illustration of the government’s strategy was the investment in broadband networks across the province. The provincial and national governments spent significant resources to build a network, in the hopes of stimulating research-intensive businesses in Brittany. Large subsidies were paid to French Telecom to develop the network, and the costs of users were subsidized. The assumption behind this expenditure was that the new infrastructure would stimulate entrepreneurial technology-oriented businesses that would be formed to take advantage of the network.

While some government-funded research centers did benefit from the spending, the anticipated benefits to entrepreneurship were not realized. As the OECD’s (Organisation for Economic Cooperation and Development) assessment concluded, “It has not yet really taken off on the business side. The indirect benefits from the promotion of broadband have not really become apparent.”²² Instead, most of the benefits from the subsidies were captured by French Telecom and publicly funded universities. More generally, while the subsidies and science parks did attract multinationals and French firms to locate there, the impact of the initiative was blunted by the difficulties in telecommunications in the early 2000s and that of the Paris-based giant, Alcatel-Lucent, more specifically.

But the French misadventures, though notable for the size of expenditures, do not plumb the depths of governmental incompetence. There are so many examples that it is hard to know where to begin. Consider, for instance, the Tenant Opportunities Program (TOP), launched by U.S. Department of Housing and Urban Development in 1988 to promote entrepreneurship. It was rolled out at 816 public housing projects between 1988 and 1997.²³ The program was not intended to promote high-technology development, but rather more modest enterprises such as handmade toy manufacturing and childcare operations.

The program in the District of Columbia was emblematic of the problems TOP encountered. It provided $2.8 million to help thirtyone complexes in the area, which were chosen through a competitive process. But while the use of the funds was supposed to be monitored by the HUD’s Washington field office, it did not have enough staff for the task. Instead, the District Housing Authority hired consultants to oversee the spending (a number of whom turned out to be related to or friends of Authority officials and members of its board of directors). Many of these consultants did not supervise the projects. In fact, a subsequent audit discovered that the majority of funds ended up in the pockets of the board members and the consultants themselves. Even as the incompetence of the consultants was revealed in the initial projects, the Authority continued to approve their use at other housing complexes.

Moreover, with no supervision and little guidance, the residents of the housing projects did not wisely spend the funds that reached them. The local councils funded efforts like Circuit City shopping expeditions and “training” trips to Las Vegas. A comment by one observer was that “word got around there was money … they knew they didn’t have to do anything for it.”²⁴

Many other parts of the United States would give the District of Columbia some competition for the hall of ignominy. Kansas would definitely be a leading candidate.²⁵ In 1970s and 1980s, the state legislature sought to boost economic development in the state. Rather than appropriating funds for this purpose, which would have meant higher taxes and angry voters, the legislature simply mandated that the Kansas Public Employees Retirement System (KPERS), the pension for the state’s employees, loan money to local businesses, and similarly to Kansas real estate developers. (The legislators somehow forgot to ask the public retirees if this is how they wanted their savings invested.) By the mid-1980s, a full 20 percent of the multi-billion-dollar pension had been earmarked for these homegrown investments.

Rather than undertaking the investments themselves, the state recruited two local investment firms to handle Kansas Investment Fund, as it was called. By the mid-1980s, frustrated at the slow investment pace, the state changed its instructions to the investment groups. Instead of backing firms that were “relatively substantial, seasoned and in sound financial condition,” they were now ordered to include new or expanding Kansas businesses that were unable to get credit elsewhere. The change of policy undoubtedly had an effect: the investment firms, which collected a fee on each transaction and had little supervision from KPERS, began putting money to work much more quickly. And they made some high-risk choices indeed. $14 million went to a manufacturer of microcomputer memories that never saw a profit, $8 million into a steel fabricating plant that soon went belly-up, and $6.5 million to a start-up that was going to develop a revolutionary hydrogen-based energy source, and so forth. The most memorable investment was doubtless $65 million in loans to a local savings-and-loan that soon thereafter was seized by regulators as insolvent. Its loan portfolio, subsequent investigations revealed, included an uncompleted Hungarian film about a man-eating bear chasing a rock-and-roll band—and a $40 million loan to the KPERS’s chairman. In all, the Kansas Investment Fund lost the state’s pensioners and taxpayers $265 million, or about 7 percent of the pension’s assets at the time. After thirteen years of litigation and $28 million in legal fees, the state recovered $41 million of those losses. (The chairman ended up being ordered to perform 200 hours of community service.)

These examples all involve government leaders who did not think carefully about realistic market opportunities and how subsidies would affect behavior. But a failure to understand the basic nature of the entrepreneurial process is also a frequent problem. One of the crucial patterns among high-growth ventures is that, at best, there are only a few very large winners. The typical outcome is disappointing. In the language of statistics, the distribution is a very skewed one. While getting comprehensive data on the returns of individual start-ups is almost impossible, we can look at the distribution of success of the venture funds that back them. Even though each fund typically invests in multiple deals, the bulk of the returns come from a few funds. Nor is the pattern unique to the United States. (Figure 4.1 shows distributions of returns of U.S. venture funds; figure 4.2 depicts the same pattern for European funds.)²⁶

Yet in many cases, government officials proceed under the assumption that success is the typical outcome. One illustration of this unwarranted optimism is the disastrous history of most loan programs to finance high-growth entrepreneurial businesses.²⁷ In many cases, governments have launched these efforts under the assumption of high repayment rates. But these programs have a fundamental issue: they do not share potential upside returns, but assume a significant portion of downside risks. For example, the Business Development Fund was established in Denmark to provide high-risk loans to high-technology projects in start-ups and established enterprises. Generous provisions for renegotiation were put in place, so entrepreneurs whose project proved disappointing were not pressured to return the money. As a result, the Fund (prior to a 2001 major reform) shared the downside risk with entrepreneurs, but received only a modest fixed interest for commercially successful projects. More than 60 percent of total funding was lost on the 900 initial projects the Fund supported.

Figure 4.1. Returns (%) of U.S. venture funds from inception to March 31, 2008

Figure 4.2. Returns (%) of European venture funds from inception to December 31, 2007

Similarly, loan guarantee programs—which transfer to the public sector part of the risk of loans to innovative firms—have a mixed track record. Again, success hinges on a program’s ability to achieve a low default rate while providing loans to borrowers that would otherwise not have been funded. Examples of loan guarantee programs include Canada’s Small Business Loans Act program, the French OSEO-Garantie initiative, and the United Kingdom’s Small Firms Loan Guarantee Scheme. The assumption that defaults will be low is frequently too optimistic: most guarantee schemes have not been sustainable without substantial subsidies. Moreover, letting a third party do the lending—most often a bank—often leads to “moral hazard” problems: the bank may be far more casual about evaluating the potential borrower when ultimately the bank’s money is not at risk. For instance, a study of the French scheme finds that the probability that a small business borrower goes bankrupt in the four years after taking out a loan goes from 9 percent if the loan does not have a government guarantee up to a stratospheric 21 percent if it does (even after controlling for the differing risk profile of the guaranteed borrowers).²⁸ Thus, a lack of foresight about the incentives government programs set up can be very costly indeed.

Capture

These tales of public incompetence seem bad enough. But much of economists’ attention has been focused on a darker problem that affects these and similar programs. Rather than worrying about government incompetence, many researchers have focused on the theory of “regulatory capture.” This hypothesis suggests that entities, whether part of government or industry, will organize to capture the direct and indirect subsidies that the public sector hands out.²⁹

This sounds terribly abstract, but it has very real consequences. A few examples from my home state, the Commonwealth of Massachusetts, illustrate the phenomenon of capture and how costly it can be:

•  In some cases, groups organize to capture subsidies being handed out by government officials. For instance, when Massachusetts governor Deval Patrick sought a billion-dollar initiative to promote biotechnology research, legislators—working in concert with local university officials—sought to ensure that funds would go to their own districts.³⁰ To cite just the most egregious example, $49.5 million was allocated to a science center at the Massachusetts College of Liberal Arts in North Adams even though the college does not have a graduate program in science. Three local three university presidents—Drew Faust of Harvard, Susan Hockfield of MIT, and Jack Wilson of the University of Massachusetts—who might be expected to be enthusiastic about public funding for research, instead criticized the bill’s emphasis on individual earmarks.

•  In other occasions, regulators create opportunities for groups to garner substantial indirect gains. For instance, Massachusetts is unique among the fifty states in requiring that a local or state policeman must be present at all road construction projects.³¹ (Everywhere else in the country, a flagman—earning only a small fraction of what the policeman does—is sufficient.) Of course, the argument is made by the police unions that this enhances public safety, but for some reason the statistical evidence does not support the claim that the presence of a policeman sitting in his patrol car munching donuts reduces accident rates!

•  Nor are these captures confined to public sector entities or employees. A law passed three-quarters of a century ago effectively limits the sale of beer and wine to liquor stores and grocery stores with fewer than four licenses statewide. Why this distinction? Once again, a powerful lobby of beer distributors and small liquor stores has blocked alterations to this policy, which allows them to charge inflated prices without competition from major supermarket chains.

Economists point out that these capture problems are not seen everywhere evenly. Rather, they tend to appear where there are individuals or firms who stand to gain substantial benefits and whose collective political activity is not too difficult to arrange. The police unions are highly vocal and march to the State House whenever the ending of police details is proposed. Meanwhile, those bearing the costs of the extra $100 million or so that ends up in police officers’ pockets each year—whether borne by a store owner seeking to widen a driveway or a homeowner paying property tax—find it much harder to undertake concerted action.

But certainly the capture of handouts is not a game that only large entities play. As Nobel laureate George Stigler points out, even very small firms can organize to benefit from public largess. For instance, industries like trucking and beauty salons have been traditionally covered by exhaustive regulations, which have made it difficult for new parties to enter the market. The lack of new competition has meant higher prices than would be the case otherwise. Despite the fact that these industries have traditionally been dominated by smaller operators, they have succeeded in getting the public sector to indirectly subsidize their profits.

If we turn to public efforts to boost entrepreneurial firms, capture problems can manifest themselves in several ways. Firms may seek transfer payments that directly increase their profits, and politicians may acquiesce to transfers to politically connected companies.

In fact some scholars have gone so far as to argue that there are two classes of entrepreneurs, some who create value, and others who simply extract profits from the system. Two closely related papers³² regale readers with stories and analyses about various characters in history—whether Chinese mandarins, medieval nobles and monks, and lawyers through time and in all continents—who were very energetic in pursuing wealth for themselves while adding little to society as a whole. Whatever we may feel personally about monks or lawyers, the papers plausibly argue that encouraging the assemblage of wealth without attention to the broader impacts on society is a recipe for disaster. (Whether we believe the statistical analysis in one of the papers, which classifies countries as those with “good” and “bad” entrepreneurship on the basis of the number of graduates with engineering and legal degrees, is another story!)

We have already seen instances of capture in the tales related above: for instance, the way in which established players like France Telecom and local universities obtained much of the funding intended to boost entrepreneurial activity in Brittany. This phenomenon is also seen in large-sample studies.

Consider, for instance, two evaluations of programs to help firms to conduct research spending. These studies were done in very different markets, high-tech Israel and Spain, which remains dominated by traditional manufacturing firms.³³ The details of the programs differed as well.

Yet despite the different economies and programs, the analyses painted a very similar picture. These subsides appear to boost research spending among small firms, in particular leading a number of modest entities to start R&D programs that would have not done so otherwise. Presumably, these firms found it hard to raise capital by other means, such as from venture capitalists or banks, and these subsidies allowed them to undertake promising projects. Among larger firms, the impact was much weaker: the public funds seem to stimulate less research, and in the Israeli case, established firms may have even cut back their own research in response to public grants.

And yet in both cases, most of the actual subsidies went to larger firms that would have performed innovative activities even had they not received the subsidy. Presumably these larger concerns, being more adept at playing the game of winning government awards, managed to commandeer the funds, even though they were much less likely to fulfill the program’s goals. The goals of the programs’ designers were thus subverted.

The phenomenon of capture also has an even darker side. In other instances, it is the organizations that are mandated to help entrepreneurs—the very ones who proclaim they are helping smaller ventures—who manage to capture much of the returns for themselves. Consider the Australian Building on Information Technology Strengths (BITS) program.³⁴ This effort was launched in 1999, with $158 million to “promote innovation and commercial success in the information industries by encouraging the creation and growth of new high technology firms.”³⁵ BITS was catalyzed by the influx of funds that Australian government obtained when it sold its 17 percent interest in the telecommunications company Telstra. The centerpiece of the program, accounting for half of the initial spending, was the creation of eleven incubator centers for small and medium-sized firms in the information technology and telecommunication sector.

The government argued that given information gaps (discussed in chapter 3), which made it difficult for them to attract investment, young Australian technology firms were not receiving enough funding. By providing financing and advice, the incubators could increase the number of small entities in the information technology business and the success of the start-up firms. In 2001, as part of a package of innovation legislation, BITS was awarded further funding. Again, in 2004, at the end of the original five-year span of the program, BITS received an extra $36 million in funding, extending the program to 2008.

These renewals might suggest that the program was an unqualified success. But two evaluations completed before the additional funding was allocated paint a somewhat different picture. In particular, they suggest that while the program had been very successful for the incubators and individuals running these incubators, for many of the entrepreneurs, the picture was not so clear.

In particular, the proportion of funds accruing to the incubators, as opposed to the firms that the legislation was ultimately trying to help, attracted criticism. At the typical incubator, most of funding went to the incubator managers themselves, in order to compensate them for management advice and other services to the entrepreneurial firms: seven of the incubators gave less than 50 percent of funding in cash to the incubated firms. One example was the incubator operated by Allen and Buckeridge Seed Stages Ventures, where ultimately a mere 31 percent of the BITS funding went to the start-ups. By way of contrast, the incubator that had the most successful entrepreneurial firms, InQbator, provided 95 percent of the funding to the companies in its portfolio.

Not only were the bulk of the funds going to the incubator managers, but in a number of cases they were actually hindering the progress of the firms. Start-ups were actually limited in their ability to shop for the best service providers: instead of finding a specialist lawyer to negotiate a licensing deal, for instance, they were forced by the incubators to use the in-house counsel (for whose services the incubator managers charged a substantial markup). The quality of the advice often did not compare with that offered by more experienced lawyers and accountants in private practice. Adding insult to injury, many of the fees charged by the incubator managers, such as for rent and telephone, were considerably above market levels. But facing the possibility of expulsion if they did not use the incubators’ services, many of the firms believed they had no choice but to agree. Unsurprisingly, more successful incubators like InQbator and BlueFire were more willing than the others to allow incubatees to obtain professional services from elsewhere.

To the government’s credit, when the BITS incubator program was provided with an extra $36 million in funding in 2004, strings were attached. The continuation of the program had stricter terms, with further funding to be allocated only to incubators that were already funded under the program and were high performing.³⁶

This phenomenon of capture is by no means unique to Australia. We’ve already alluded to venture promotion programs, such as the SBIC program in the United States, that continue to exist long after they’ve exhausted their usefulness. The presence of a vocal “subsidy lobby”—typically, trade associations representing groups that are benefiting far more from the subsidies than the entrepreneurs the programs are designed to help—is typically the root cause.

FINAL THOUGHTS

The final pillar on which the case for public intervention in venture activity rests is the claim that public intervention can be effective. Unfortunately, this argument is undoubtedly the wobbliest.

Certainly, entrepreneurial and venture capital activity exhibits many of the same features as other activities that receive public subsidies:

•  It is much easier being an entrepreneur when one has many peers, which makes the task of initial pioneers particularly hard.

•  The government can presumably provide certification to little-known entities.

•  The knowledge generated by any one venture is likely to benefit many others.

But the same pathologies that bedevil many government efforts to provide subsidies are likely to emerge when the target is entrepreneurship. Entrepreneurial activities are by their nature highly uncertain and unpredictable, which means errors are common. And this uncertain environment means that those who want to direct subsidies to themselves may be able to operate with little scrutiny.

In the next four chapters, we’ll turn from the “thirty-thousand-foot level” where we have been dwelling to a much closer look at the design and implementation of investment programs. The chapters will seek to understand what common mistakes governments make when putting these programs into practice, and how they can be avoided.