Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
The Roles of Technological Start-Ups and Small Innovative Companies in the U.S. Economy Commercial innovation is inherently risky. To the normal busi- ness risks, technical innovators add the risks of technical failure or failure to match a technological advance with a market demand. in certain types of commercial technological advance, exploration by small firms is a critical part of the learning process of the indus- try. This is not a new phenomenon. in the early days of the automobile industry, for example, the development of technology and defacto industry standards were driven by small technological risk-takers. Between 1895 and 1923 the number of automobile manufacturers grew from O to 75. in 1923 Docige introduced the all-steel, closed-body automobile and the number of companies dropped precipitously as this new industry standard (and a much smaller number of companies) grew to capture 80 percent of the market by 1926. The scale advantages in production were reflected in sharply lower costs, and by the early 1930s the renumber of auto- mobile manufacturers had dropped to 25.~ Historical distance al- Tows us to see this pattern clearly, a pattern that future historians AM. Utterback, Innovation and industrial evolution in manufacturing indus- tries. Pp. 16-48 in Technology and Global Industry, B. Guile and H. Brooks, eds. (Washington, D.C.: National Academy Press, 1987~. 15
16 RISK AND INNOVATION of industrial technical change are likely to be able to identify in personal computers, software, and biotechnology. Tncleed, the biotechnology industry represents a particularly dramatic contemporaneous example of the phenomenon. The for- mation of this new industry was catalyzed by advances in funda- mental biology in the 1960s and 1970s, and in particular, by the invention of technology that provided the capability of synthesiz- ing complex natural proteins, such as human insulin, by inserting the requisite gene into simple microorganisms. The rate of forma- tion of new firms dedicated to the exploitation of one or another aspect of this new technology has been phenomenal; approxi- mately 800 new enterprises were founded in the decacle of the 19SOs, and the industry currently numbers more than 1,200 firms. A few of these firms have become large, successful operating com- panies (e.g., Amgen), but the vast majority of biotech firms are small investor-funded R&D ventures. The biotechnology industry has demonstrated itself to be an effective vehicle for rapid societal exploration of the potential of a set of powerful new technologies. However, it appears that the associated high risks were somewhat obscured by a few dramatic early successes achieved by the pioneer firms. Most recently, the spectacular advances in biotechnology have become qualified by the high costs for a course of therapy that uses the new drugs, reflecting the high spending by biotechnology companies to de- velop these new drugs. A recent spate of failures of prospective biotech medicines in clinical trials, couplect with the emergence of a more difficult market for costly health care products, has made raising capital extremely difficult for all but the best-positioned biotech firms. A closed financial market, if long continued, could lead to a substantially restructured industry, and perhaps the clos- ing of this "exploratory" phase of biotech commercialization. These demancls of pioneering new markets of driving such societal technological learning change dramatically in short peri- ods of time as a function of changes in technology, market de- mand, and industry maturity. Table ~ names a few of the typical factors increasing and decreasing risk borne by companies, of all sizes, engaged in technology-dependent ventures. Some of these risk factors are best negotiated by larger companies-they may be better equipped to minimize the risk or survive an adverse out- come but some are successfully tolerated or managed only by
TECHNOLOGICAL START-UPS AND SMALL INNOVATIVE COMPANIES 17 TABLE ~ Factors that increase and Decrease Risk Associatecl with Technology Investments Factors Increasing Risk Factors Decreasing Risk Totally new market or low experience in the market with the product or service Strong competitors Technological uncertainty Environmental uncertainty High potential product liability (medical products, nuclear, tonics, etc.) Changing industry standards Marginal internal skills- not "leading edge" Restricted market access, especially in worldwide markets Little protection for intellectual property Regulatory barriers to commercialization Expansion in existing market No dominant competitors Government technology funding or steps to create the market by purchases Stable standards environmental, technological, social, etc. Relevant government infrastructure External investment partners with large resources (cooperative venture) "Safe harbors" from product liability for certain products (vaccines, defense products, etc.) Guaranteed access to foreign or government technology or other external sources Strong patent or copyright protection small companies. Small companies may systematically face a dif- ferent risk/rewarcl line than larger companies; some risks may be differentially critical to small companies. For example, small suppliers to large companies may have to commit the bulk of their company's resources to satisfy the de- mands of a single buyer. For such a small company, the conse
18 RISK AND INNOVATION quence of a change in plans or purchasing by its primary customer, or of a failure on the part of the small company to deliver quality product, could easily be bankruptcy. Another example is a policy or regulatory change that lengthens the time to market for a new product being brought out by a start-up company. The costs of keeping the company solvent during the lengthened period be- tween development expenditures and positive cash flow may be too much for a thinly capitalized small company. In a larger com- pany, it is much more likely that the resources will be available to tolerate a clelay. COMPARING THE CONTRIBUTIONS OF SMALL TECHNOLOGY-ORIENTED COMPANIES The technological risks of commercial companies are different than those of non-market research and development. In adclition to the risk that a new technology may not work as expected, there is the competitive market risk: can a product or service using a new technology cleliver adequate value for its price in comparison with competing offerings? In general, to reduce real and persistent tech- nical and market uncertainties, the bulk of small technically ori- ented companies and start-ups tend to focus on incremental market- driven innovations, and not breakthrough technologies; small company innovation often reflects the shape of perceived market opportunity. Based on the industries studied for this project, it is clear that small companies are often compelled to focus on incre- mental, market-driven innovation by the immediate, anct at times overwhelming, risk of running out of money. Further, the limitations of small companies' abilities to reduce technical and market uncertainty constrain the character of new opportunities and drive dependence on outside technical resources. For most small companies the opportunities for incremental com- mercial innovation are shaped by advances in science and technol- ogy from outside the company usually in universities or large company research laboratories; often entrepreneurs are intimately familiar with these advances, either as they were directly involvecT or know well the people doing the research. The ability of small companies to develop, in-house, advances in science and technol- ogy that create new opportunities is usually restricted to a rela
TECHNOLOGICAL START-UPS AND SMALL INNOVATIVE COMPANIES 19 lively narrow scope, reflecting the expertise of the technical mem- bers of the small company, as well as their previous experience. Government policies also shape the opportunities and the level of risk for small companies, both directly and indirectly. Direct government funding research grants and government mission- oriented R&D contracts- can recluce the risks of technology com- mercialization for small companies. These types of support are most important for emerging technologies that have not yet been proven to be commercially viable. Much more frequently, how- ever, government policies shape the opportunities for small com- panies indirectly. in particular, small high-tech companies tend to be relatively vulnerable to risk created by the legal system and by government policies such as health and safety regulations, em- ployment and environmental regulations, and export controls. The industries studied for this project illustrate the types of technological and policy risk borne by a representative group of small companies. Most important, they illustrate how small com- panies pioneer and develop new markets and provide product di- versity and innovation in small markets.2 Advanced Displays and Visual Systems Although cathode ray tube (CRT) technology continues to dominate electronic displays, large and expanding opportunities exist where the CRT's power requirements and physical dimen- sions cannot be accommodated, and alternative display technolo- gies must be used. These technologies include increasingly ubiqui 2The brief industry descriptions provided in this chapter are drawn from the longer industry studies prepared as part of the study project. The estimates of typical or average company size, market sizes, and market growth rates in each of the industry descriptions are drawn from an unusually diverse set of pub- lished resources, most often the industry's trade literature or material prepared by the industry's trade association. The arguments about industry structure and dynamics were developed during the industry-specific workshops held for this study or drawn from trade and industry association publications. The sources for these generally reliable but less-than-perfect numbers and arguments are pro- vided in the separately published industry studies. National Academy of Engi- neering, Small Companies in Six Industries: Background Papers for the NAE Risk and Innovation Study (Washington, D.C.: National Academy Press, forthcoming in 1996~.
20 RISK AND INNOVATION tons flat pane! displays, already used in full-color notebook com- puters, aircraft avionics, and a variety of handheld computer games and electronic devices, as well as new display technologies that typically involve projection for presentation systems and wide- screen high-definition televisions. These advanced displays are expected to be among the most critical and perhaps expensive com- ponents of the next generation of these devices. For example, the market for just flat panel displays is expected to triple by 2000, from $3.7 billion in 1993. Advanced displays and visual systems are unique in the sec- tors stuclied for this project, in two respects. First, Japan dominates advanced displays, not the United States. In 1992 Japanese compa- nies produced 93 percent of the world supply of flat pane! dis- plays, and essentially built all screens using active-matrix liquid crystal display (LCD) technology. Competition among the major Japanese companies has accelerated Japanese investment in this technology to more than $3 billion by 1993, effectively making 1:CD the dominant advanced display technology for the foreseeable fu- ture. This is in contrast to the situation that has evolved in the United States, where there has been an almost complete absence of large investments in active matrix LCD production to date. The second respect in which advanced displays and visual sys- tems are different from the other sectors in this project is in the relatively few U.S. companies that compete in advanced displays. In contrast to Japan, efforts to commercialize advanced displays in the United States over the last decade have been largely by a group of 10 to 20 small, technology-intensive companies pursuing a vari- ety of next-generation technologies. Many of these are spin-offs from the R&D efforts of larger, technology-based U.S. companies that exited the advanced display market in the 1980s. The reason these small companies exist in advanced displays is that, with few exceptions, the market pull for the next generation of display technologies is undeveloped in many potential applica- tions. For example, plasma and electro-luminescence technologies compete with LCDs, as well as with several other emerging tech- nologies (e.g., field emission displays, microtip displays, vacuum- fluorescent displays). But no one technology has yet emerged that is able to meet the requirements (e.g., cost, power consumption, image contrast, color, manufacturability) for all applications. This creates enormous technological uncertainty that centers on a few,
TECHNOLOGICAL START-UPS AND SMALL INNOVATIVE COMPANIES 21 critical questions: How well will the new technologies perform? How expensive will these technologies be to develop and produce in volume? Will these new technologies be able to eclipse current LCD technologies that are likely to continue to improve? in response to this uncertainty, small U.S. companies have ex- ploited opportunities to develop different technologies for differ- ent applications. in this respect, they do not conduct research per se; rather they focus on technically challenging product develop- ment and engineering. Few if any companies have grown solely from technological or scientific breakthroughs. The most success- fu! small companies have identified a market need anci aclapted existing technology to fill that need. Federal technology policies, in particular support from the Advanced Research Projects Agency of the Department of Defense, have also been critical to developing these technologies by creating a degree of stability in the opportunities for small U.S. display companies, typically by awards directly to small companies in the form of grants or development contracts from the Department of Defense. More recently, mergers and partnerships between these small entrepreneurial companies and larger companies have also become increasingly frequent, for several reasons. As opportunities for acivanced displays have been proven (primarily by Japanese com- petitors), larger U.S. electronics companies have realized (a) the large current and potential markets for the right new technology, (b) the value of in-house capability in components as fundamental as displays, anc! (c) the risks of not being current in display product and production technology. Small it&D-intensive companies are increasingly aware that the challenges and cost of scare-up from prototype to production effectively limit their ability to push technology. First among these challenges is obtaining sufficient capital. This is especially true of LCDs for which a single factory to produce displays in volume can cost between $100 million and $300 million. In addition to capital, there are a number of serious technical challenges associated with the manufacture of displays. For example, a lack of standardized process equipment greatly complicates display manufacturing. The United States has also lacked a robust display infrastructure in which to develop, test, and sell equipment. The supply of properly trained scientists and engineers has also been small.
22 RISK AND INNOVATION In summary, small companies in advanced displays play an unusual role. The demand for advanced displays is expected to be very large, but many of the markets are (as yet) unrealized, creat- ing enormous uncertainty about the next generation of display technologies. This technological uncertainty and therefore risk- as well as direct government support of advanced displays tech- nologies has created a cadre of small, relatively research-intensive companies in search of a valuable proprietary technological posi- tion. These companies represent, in effect, a repository of technical capability in advanced displays. But the capital requirements for scale-up and the technological issues of manufacturing for large markets- including the relatively weak infrastructure in the United States for advanced displays severely limit the opportunities for small companies to exploit these technological advances. indeed, they will increasingly be compelled to form partnerships with larger companies or be acquired, as core and complementary tech- nologies in displays advance. Implantable and Surgical Medical Devices The U.S. market for medical devices and dental equipment is estimated at around $40 billion. This significant aggregate figure, however, conceals a fragmented market including on the order of 1,700 different types of medical devices. The term "implantable and surgical" refers to the subset of medical devices intended for implantation in the human body, such as shoulder prostheses and left ventricular assist devices, or for manipulating human organs and tissues, especially devices used to perform minimally invasive therapy. Some of the more impor- tant devices include angioplasty catheters, endoscopes, and a vari- ety of accessory device technologies including lasers and miniatur- ized forceps. The size of the market opportunities for these high- tech devices tends to be relatively modest, ranging from tens to hundreds of millions of dollars per year in the United States. Within vascular surgery, for example, the size of the market in 1991 ranged from less than $:~0 million for vascular probes, to about $30 million for carotid shunts, to about $70 million for vascular grafts. Technological innovation in these niche-markets depends criti- cally on this risk-taking by entrepreneurial individuals and small companies. in contrast, larger companies may not have the ability
TECHNOEOGICAL START-UPS AND SMALL INNOVATIVE COMPANIES 23 or patience to develop innovations for small markets the risk is too high and the return too Tow. The economics of device research, development, and demonstration can make these markets unat- tractive to larger companies until the application of technology is well proven in the market. The fragmentation in these markets is reflected in the relatively large number of small companies that are involved in mectical crevices between 10,000 and 11,000 device firms, either domestic or foreign, operate in the United States. Of these, more than 70 percent of U.S. medical crevice manufacturers have fewer than 50 employees. Universities are important to medical device innovation in the United States. First, universities and academic medical centers tend to be the locus for innovation in medical devices, as well as for the development of new devices. Frequently, medical device inno- vation is driven by what can best be characterized as "hobby shop" innovation by highly trained specialists in universities or academic medical centers. in particular, devices used in limitect quantities- such as specialized stents used to hold open clogged vessels-are often designed and developed (at least initially) by individual sur- geons or physicians who, as part of their academic or clinical work, see a better way of doing something and work with a small team to create a new device or paired device and procedure. Universities are especially important in another sense. In con- trast to pharmaceutical development, innovation in implantable and surgical medical devices tends to look far outside the realm of biomedical research for new technological opportunities. Medical device innovation results largely from incremental, engineering- basect innovation that transfers new technologies into medicine. Innovations in imaging technologies are an example of this. The computerized tomographic (CT) scanner or magnetic resonance imager (MRI), for example, depended on the transfer into medicine of advances in other areas, including mathematics and computers, electronics, optics, ancT material sciences. Equally, much of the technological dynamism of current developments of the encloscope use semiconductor chips for imaging, as well as fiber-optic tech- nologies. Universities and academic medical centers are especially important as the locus of this technology transfer. Government policies also have a powerful impact on medical device innovation. First, the government is a major source of medi- cal R&D funding. Second, the Food and Drug Administration
24 RISK AND INNOVATION (FDA) influences the selection and development of new medical crevices through its premarketing approval and regulations for medical devices. Third, the government, primarily through the growth of Medicare, has become a major source of payment to the providers of medical services. Consequently, government deci- sions about what to pay for, and how much to pay, are an impor- tant influence on the incentives for small company risk taking. Currently, radical changes in this regulatory and policy envi- ronment for medical devices are raising questions about the viabil- ity of the innovative enterprise which has driven linked technical and therapeutic advances in smaller segments of the medical de- vice industry. First, more stringent FDA regulation of mectical devices requiring increased numbers of trials and evaluation in- creases the expected time to market of new devices and increases the cost of demonstrating new devices, increasing the business risk. This translates directly into a more difficult financial environment for device start-ups. Second, a serious effort continues toward radical reform of the health care financing and delivery system through legislation. These reform efforts and changing reimbursement policies also in- crease the uncertainty with regarc7 to device markets anct conse- quently the risk. Increased sensitivity to the cost of medical care has also changer! the mocle! for adoption and use of new medical crevices. Previously, adoption in medical devices depended on an intimate mocle! of evaluation and dissemination of new medical devices by colleagues of the inventor and other physicians; adop- tion decisions were not particularly sensitive to cost. By contrast, a different model is now emerging as payers health care insurers and federal agencies increasingly restrict, through decisions af- fecting cost coverage, the adoption of new technologies by indi- vidual physicians. These changes may have vast implications for innovation and for the viability of small companies in medical de- vices. To summarize, small companies and entrepreneurs are exceed- ingly important for innovation in medical devices. The high de- gree of regulation in medical devices, as well as recent changes in this regulation, affects the economics and direction of medical de- vice R&D, creating a high degree of economic risk. This risk, as well as the large technological risks and relatively fragmented mar- kets for medical devices, makes these opportunities unattractive to
TECHNOLOGICAL START-UPS AND Salt INNOVATIVE COMPANIES 25 larger companies until the technology and application have been demonstrated. For these reasons, medical device innovation and product diversity appear to depend, perhaps to a surprising ex- tent, on individual user-inventors. in an academic or clinical set- ting usually a university or academic medical center these user- investors transfer technologies developed in other areas and for other uses into medicine and in that way create new medical de- vices. Software While the roots of the software industry are firmly planted in the early history of commercial mainframe computers, the current software industry largely unbundled from hardware and domi- nated by applications running on personal computers, worksta- tions, and networks did not exist as recently as 15 years ago. In economic terms, the majority of software development continues to be done internally by companies spending for maintenance and incremental improvement to existing software is estimated to be as high as $:~50 billion to $200 billion. But spending for new software is increasingly used to buy either prepackaged software or customized software and services. This spending for new soft- ware and services is estimated to exceed $100 billion worldwide. Despite the increasing number of large software companies, this high level of spending for new software continues to create an extraordinary number of start-ups and small software companies. Indeed, in 1987 more than half of all U.S. software companies had fewer than five employees, 97 percent had fewer than a hundred employees, and this level of concentration in software has declined in the last clecade. Small companies thrive in the software industry, primarily be- cause of two trends: (a) the rapid pace of change in electronics anct software technologies, and (b) the high degree of fragmentation in markets as electronics and software seep into almost every aspect of the economy. in other words, despite its large economic size, demand for software in established applications (those a few years old) continues to increase even as technological advances and hu- man inventiveness create new market opportunities such as multi- media. This fragmented, rapidly growing marketplace is an ideal envi
26 RISK AND INNOVATION ronment for start-ups that exploit technologies or markets deemed to be too small or too risky for established players. Thus, new markets and narrow, niche markets that lead to considerably larger markets let new software companies develop the revenue stream, product, and core competencies of valuable new software busi- nesses. The nature of software development also means that initial bar- riers to entry are extremely low although in prepackaged soft- ware the costs and the distribution problems of selling into high- volume markets are increasing the capital requirements for new entrants. Nor clo there appear to be many economies of scale in cleveloping these applications. Indeed, the imperatives of creativ- ity and flexibility in software clevelopment appear to favor innova- tive, new competitors seeking to serve limited or specialized mar- kets. These dynamics make the investment risk-reward equation for many untested opportunities too demanding for large software companies. Insteacl, larger companies usually commercialize new technologies and enter new markets by acquiring a smaller com- pany or start-up that has demonstrated success. This mechanism creates both substantial industry growth and opportunities for en- trepreneurs to reap returns on their investments. This economic organization of the software industry has conse- quences for the viability of new software companies. New compa- nies may identify new markets or new technologies, but they are extremely clependent on external sources for access to capital, mar- kets anci crucially, for distribution- management expertise. They also quickly become vulnerable to new companies with a better idea. As a result, software companies tend to be divided into three groups. The first group consists of the few companies that become large and develop the internal resources to have long-term staying power. Examples are Microsoft and Lotus. But these are exceptions. The overwhelming majority of start- ups in software are in the second group, which develops niche- market products and reaches revenues in the $5 million to $15 million range. The life cycle of these companies is short relative to older industries typically they will either fail when their product life cycle has run its course or be acquired by or merged with other companies to reach sustaining capabilities. The third group in- cludes those software companies that for a variety of reasons are not successful, and fail. .
TECHNOLOGICAL START-UPS AND SMALL INNOVATIVE COMPANIES 27 Finally, in software, uncertainty over the definition and protec- tion of intellectual property rights also has an important impact on the opportunities for small companies. Patents are expensive to obtain and are difficult to enforce and defend. in this sense, the economics of litigation favor larger companies at the expense of smaller ones. There is also ambiguity about what is and is not patentable. These ambiguities also increase risk by increasing the costs and uncertainties of litigation. Small companies in software, then, represent in a nearly ideal form the entrepreneurial, technology-intensive start-up. Although the number of large competitors in software is increasing, the rapid development of new technologies and enormous number of new potential markets create a broacT array of opportunities for techni- cally capable small companies. The range and number of opportu- nities, low barriers to entry, and absence of significant economies of scale in the development of many new specialized software products favor innovative, new companies. indeed, to a consider- able extent, development and commercial proof of emerging tech- nologies and markets in software depend on the investments and actions of small companies willing to accept technological (and financial) risk. l Environmental Testing Services The roots of the environmental testing industry can be traced back to 1962, when drinking water standards advancect by the U.S. Public Health Service spurred the development of water quality testing labs. This nascent industry consisted of small in-house water utility labs, as well as commercial "mom and pop" laborato- ries that tested for a variety of simple parameters such as water hardness, some metals, and coliform bacteria. Industry growth accelerated with the introduction of the first two major pieces of environmental legislation requiring extensive analytical testing- the Federal Water Pollution Control Act amendments of 1972 (now referred to as the Clean Water Act, or CWA), and the Safe Drinking Water Act of 1974 (SDWA). A second wave of major environmental legislation increased the demand for commercial environmental laboratory services ex- ponentially starting in the late 1970s and early 1980s. The two seminal pieces of legislation were the Resource Conservation and
28 RISK AND INNOVATION Recovery Act of 1976 (RCRA), which regulates the treatment, stor- age, disposal, anct transportation of both hazardous and nonhaz- ardous solid wastes, and the Comprehensive Emergency Response, Compensation and Liability Act of 1980 (CERCLA, also known as Superfund), which requires the cleanup of abandoned hazardous- waste sites. With the introduction of RCRA and CERCLA, and subsequent reauthorizations and amendments to all the major en- vironmental laws, the commercial laboratory industry grew rap- idly. Most observers of the environmental testing industry believe that the industry began to plateau starting in 1988 or 1989. Growth rates averaging 25 percent or more during the mid-1980s have slowed to around 5 percent currently. The current size of the commercial market for environmental testing services is probably between $~.5 billion and $~.6 billion, with annual revenue growth being less than 5 percent. Testing under RCRA accounts for about half of total industry revenues, Superfund (CERCT~A) another 20 percent, CWA 15 percent, and SDWA 5 percent, with the balance being divided among various other environmental statutes. Estimates of the current U.S. envi- ronmental instrument market are about $500 million. The number of environmental testing laboratories in the United States is between 1,400 and 1,600, virtually all of these small. Of these, only 25 to 30 environmental testing labs generate revenues greater than $10 million. This is because market demand is gener- ally local (markets are small) and economies of scope and scale, which could enable larger companies to compete at lower cost, are not eviclent. While there are many small testing laboratories be- cause of low barriers to entry, the proven lack of economies of scale and scope limit their growth prospects. In contrast, the equipment industry that serves the testing labo- ratories is markedly less fragmented anc! is a global opportunity. The top three environmental instrument manufacturers Hewlett- Packard, Perkin-Elmer, and Thermo Instrument Systems- together account for almost 30 percent of environmental instrument sales globally. Small instrumentation manufacturers, of which there are probably several dozen in the United States, tend to specialize in niche markets (for example, waste water samplers) or emerging technologies (for example, field testing technologies). Because the market for environmental testing services is cre- ated primarily by regulation and driven by enforcement, the entire
TECHNOLOGICAL START-UPS AND Sail INNOVATIVE COMPANIES 29 population of laboratories is subject to a substantial degree of non- technological (and nonmarket) risk specifically, the uncertainty inherent in the fits and starts of a political and regulatory process that effectively defines their market and determines their growth rate. Equally, the inflexibility inherent in these regulatory schemes decreases the incentives for environmental testing labs and instru- ment suppliers to be innovative technologically. indeed, environ- mental testing labs are unique in this respect, in that it is the only sector studied in this project where technological risk is almost entirely absent. In many cases regulations preclude proprietary "product" technologies; regulation of testing processes reduces technological uncertainty in the market for services and limits the potential benefits of small company innovation. As a consequence, current technical innovation in environmental testing labs focuses on incremental approaches to cost savings and throughput in- creases rather than new product development, which is largely irrelevant. in contrast to environmental testing labs, some forces clo continue to drive innovation by instrument suppliers. There is an active R&D effort in new technologies and new applications of existing technologies to enable labs to reduce cost and improve flexibility through (~) improved information management systems, (2) increased automation, anct (3) increased use of portable "field testing" technologies. As well, as the number of regulated com- pounds increases, the limitations of existing methods for various analyses may require the increased application of newer technolo- gies, such as high-pressure liquid chromatography, liquid chroma- tography linked with mass spectroscopy, and sequential mass spec- troscopy although none of these technologies, for the most part, represents "cutting edge" science. In summary, although environmental testing labs are techni- cally oriented, demand for environmental testing services is cre- ated almost entirely by regulation and driven by enforcement; without these labs, aspects of current environmental regulation simply are not possible. This, in turn, introduces a high degree of risk for these labs, inherent in changes to this regulatory scheme. Consequently technological innovation tends to be incremental and to focus on cost savings. The overwhelming majority of environ- mental testing labs are also quite small. This reflects extremely low barriers to entry but also barriers to growth no marked econo
30 RISK AND INNOVATION mies of scale have yet been demonstrated by environmental testing labs, and demand tends to be local. in contrast, the equipment industry that serves the testing labs is considerably less fragmented and is the locus of most R&D expenditures for the industry and the source of most process innovation. Network Services and Access Devices The consent decree that broke up the Bell system in 1982 had profound and far-reaching impacts in telecommunications by deregulating the long-distance telephone business. in parallel, that decade witnessed revolutionary improvements in computer tech- nology, fiber-optic communications, and ability to communicate digital information at high rates on copper wires. These events in combination ushered in the "information age" in which we live. Over time, we have seen the merging of the wide area and local area networks and the expansion of other communications media with wireless ground and satellite-based technologies. As a result, enormous opportunities were opened for successful new busi- nesses, and many companies have sprung up providing services or equipment or software (or a combination) in the general industry categories broadly denoted "telecommunications," "data commu- nications," and "networking." Among the earliest beneficiaries of deregulation were a host of new long-distance telephone service providers who, with modest capital, were able to underprice AT&T. After some years, these companies were acquired and consolidated into larger entities, such as LC} anct EDDS, competing in the same markets with giants such as MCT (an early start-up), Sprint, and AT&T. Starting such a long- distance carrier is no longer a realistic business opportunity as costs favor the large players and pricing has become extremely competitive. On the device and system side, new companies grew up providing network access products and related software. Telco Systems, Tnc., Newbridge Networks Corporation, Cisco, 3Com Corporation, Level ~ Communications, and SynOptics are ex- amples of such successful companies, but there are many others. Note that these companies thrived despite the presence of estab- lished computer companies in the network equipment market, such as IBM and Digital Equipment Company, by providing new procl- ucts tailored for emerging market segments.
TECHNOLOGICAL START-UPS AND Sail INNOVATIVE COMPANIES 31 . Opportunities in these industries are being pursued by compa- nies of all sizes that are capable of creating or combining software and physical assets to create networks and deliver services over those networks. This industry's fuzzy boundaries, its overlapping and competing technologies, its many participants, and its rapid growth rate make meaningful estimates of total size impossible, but it is very large. As indicators it is worth noting that annual U.S. local phone service revenues are about $80 billion, cable TV rev- enues exceed $20 billion, and "small" equipment markets such as the market for video conferencing equipment-are conservatively projected to grow at 20 to 30 percent for the next several years. The dramatic growth of networks both fuels and is fueled by dramatic fragmentation of opportunities. Much of the incremental development of networking technologies (as opposed to funda- mental advances, such as optical transmission) required to realize new opportunities is done by new companies or entrepreneurs. in this industry small companies fill important niche markets and explore market acceptability for products considered too risky for larger companies: · New companies are expected to play a key role in the viabil- ity of portable computing. The number of people using cellular and satellite systems to transmit data is anticipated to increase from less than million users in 1993 to more than 20 million users in the next decade. These broadband, interactive video and data ser- vices especially are seen as an opportunity for new companies, because the demand for these services is unrealized. This capabil- ity is also expected to create demand for new wireless devices such as keypads, integrated devices for voice and data, and future gen- erations of personal digital assistants (PDAs) like Apple's Newton which link cellular, wireless fax, and e-mail. · Satellites are a spectacular demonstration of advance in in- formation technology, creating the opportunity to provide instan- taneous point-to-point communications or broadcast to anywhere on the globe. While satellite networks are mostly a large company proposition, the terminal devices for many satellite services, in- cluding positioning and navigation devices, and the personal digi- tal devices needed to receive facsimile and data are expected to be developed by new companies at least initially. The terminal de- vices for the Global Positioning System, or GPS, for example, have been developed largely by small companies.
32 RISK AND INNOVATION · The demand for broadband services, including full-motion vicleo or multimedia, is creating demand for new, faster network- ing technologies. Existing local networks such as Ethernet, token ring, and fiber distributed data interface (FDDT) will eventually make the transition to faster technologies, such as asynchronous transfer mode (ATM) or "fast Ethernet." Implementation of these new broadband network technologies is creating enormous inno- vation in network equipment vendors, many of these by small or new companies. Uncertainty about Information network architectures and stan- ciards shapes the opportunity for start-ups in network devices. Because standards are in flux, it is still possible for a small com- pany to set a de facto industry standard. Network devices and services are embedded in the telecommunications industry, an in- dustry that has historically been heavily regulated for example, in spectrum allocation, antitrust, and pricing. In this context, the absence (often temporary) of standards, Including even proprietary standards, has encouraged innovation in network devices. This has created opportunities for small companies that are willing to accept a high level of risk, and simultaneously has discouraged larger competitors. In summary, a high degree of commercial and technological uncertainty is forcing the larger players in networking to engage in a frenzied competitive search for solutions that simultaneously cre- ate de facto technical standards and establish the proprietary net- work technologies that will dominate the nation's information in- frastructure. in this competitive environment, larger competitors depend on small companies to realize new opportunities that will drive demand for (a) communications capacity in these new net- works, as well as (b) the core technologies such as compression technologies and portable computing. Outdoor Sporting Goods The $4 billion to $6 billion U.S. outdoor sporting goods indus- try including backpacks, climbing ropes, kayaks, tents, and ski parkas, to name a few items is nested within the larger industry of general sporting goods. These sporting goods markets are domi- nated by small privately-helc! companies- perhaps 5 percent are
TECHNOLOGICAL START-UPS AND SMALL INNOVATIVE COMPANIES 33 public companies, and only six of these have a market capitaliza- tion of over $l billion. Many of the markets these companies serve grow rapidly with the growth of a particular sport; it is not un- usual for sport participation to grow rapidly for several years. For example, in the early 199Os participation in two sports rock climb- ing and in-line skating grew at more than 20 percent a year for several years in a row. The outdoor sporting goods industry especially in "hot" sports is characterized by innovation-oriented start-ups. The costs of entry are fairly low, opportunities for the application of new technology (often new materials technology) exist, and mar- kets are fragmented and often small enough not to be of interest to larger companies. Further, the outdoor sporting goods industry is one in which user-inventors and trendsetters traditionally play a large role. In outdoor sporting goods, a climber, hiker, or paddler envisions a product he or she would like to have, creates a proto- type in the basement, and manages to grow a company on the strength of a product that others, with similar experience to that of the inventor, appreciate. The result is that many companies in the industry have a strong predisposition to plan on product innova- tion as part of their company strategy. This is in marked contrast to process innovation, which tends to happen only in response to product and volume requirements. Innovation in the outdoor sporting goods industry is typical of consumer product industries where the line between changing fashion and advancing product technology is usually blurred. Tn- novation in outdoor footwear is an excellent example. In the past clecade hiking boots have become lighter while becoming more waterproof, easier to break-in, and easier to care for. Following the trend in athletic shoes, outdoor footwear has been revolutionized by new materials. Anyone who has ever broken in a old-style pair of heavy leather hiking boots will immediately recognize the tech- nological advances inherent in the new, lighter generation of foot- wear. The dramatic increases in sales of the product, however, cannot be attributed simply to the technical superiority of the new boots. Lifestyle changes- an increase in the number of people walking on forest trails for recreation-are important. Even more important to the growth in sales are fashion changes; industry watchers are aware of the shift from "white" to "brown" shoes in the preferences of U.S. junior high students. Similarly, "amphibi l
34 RISK AND INNOVATION Gus" sandals designed originally for rafting and other water sports have evolved from clumsy special-purpose shoes into mul- tipurpose outdoor footwear. While there is technical advance in the construction of the sandals, it is nonetheless an evolution driven by fashion. Similarly, while rollerskates are an old product, the rapidly growing sport of in-line skating ("blacles" rather than "skates") was created out of innovation in materials and industrial design that allowed people to skate smoothly on concrete and as- phalt. in short, innovation in many outdoor sporting goods is most important, and most prevalent, when it reinforces or responds to lifestyle and fashion changes. Companies in the industry remain technically innovative largely without much organized research, materials testing, prod- uct clevelopment, and product testing-by relying heavily on a small cadre of designers and innovators. A few "product design- ers" in these companies make virtually all decisions about aesthet- ics, structural characteristics, manufacturability, raw material specs and purchasing, pricing, testing, presentation, and shipping. The concentration of product and manufacturing process development in a few product designers also affects the organization of market research. Companies count on the personal outdoor experience of a few designers with the products to stay close to the market. Simi- larly, they count on the creativity of the same group to bring new technology (usually first developed outside the industry) to bear on either products or manufacturing processes. in other words, the size of the companies, and of the markets most companies serve, pushes to keep the product development costs low and the various design functions centralized. At the same time the technical characteristics of the products are such that it is possible for a single individual, or a couple of individuals, to execute most or all of the design functions. The "personal" nature of most of the products, and the fact that many designers are out- door enthusiasts, allows product development decision making to be centralized in people close to these markets. AcIditionally, ex- ecutives in the industry refer to the necessity of maintaining ad- vanced design and technology as well as a "feel" for the products and the industry. Many of the small companies in the outdoor industry are best characterized as lifestyle companies. A single founder or small team of individuals creates a company (or buys an existing com
TECHNOLOGICAL START-UPS AND Sail INNOVATIVE COMPANIES 35 party) out of commitment to the sport and an interest in being a long-term participant in the industry. In sharp contrast to software start-ups, for example, many small companies in the industry even the most innovation-oriented companies do not expect to grow rapidly or to go public or be purchased by a larger company as a method of getting founders "liquid." Technical entrepreneurs in the industry are drawn primarily from the ranks of enthusiasts who want to be a part of the industry rather than from general entrepreneurs who see the industry as ripe for innovation-based businesses. Finally, among the industries studied for this project, with the possible exception of prepackaged software, this industry is most dependent on established distribution and retail networks. Retail consolidation may increasingly squeeze small manufacturers as some retailers attain the volume necessary to integrate backward profitably into production; innovation may shift away from prod- uct design toward production cost savings. In summary, the outdoor sporting goods industry is typical of a consumer product industry clominated by small companies. There is a considerable degree of technical entrepreneurship, the impacts of which are often indistinguishable from the impact of lifestyle and fashion trends. The industry is fragmented, with sub- stantial pockets of technical dynamism and rapidly growing mar- kets, but the expectations of entrepreneurs, and therefore the shape of the companies they build, are different from those in industries such as software or network devices. Summary Comparison The industries addressed in this study illustrate the diverse roles that small technically oriented companies play in an industry and in the economy. Small companies in the software and the network devices and services industries are prototypical entrepreneurial high-tech com- panies. New technologies and new potential markets seem to cre- ate a wide range of opportunities in these industries for technically capable risk takers. Also, new company starts are supported by (a) relative ease of entry because of relatively Tow initial capital re- quirements; (b) a significant number of public business successes to encourage entrepreneurs to enter the industry; (c) companies
36 RISK AND INNOVATION that appear to have substantial value if sold, even though they are not necessarily long-term survivors; (~) availability of venture capi- tal financing; and (e) a public market for the shares of companies at relatively high valuations, compared with other industries, on the basis of revenues or profitability. in both of these industries there are large, dominant companies, but the technological dynamism and fragmentation of the market create opportunities for small companies. In both software and network industries the devel- opment and commercial proof of a wide range of emerging tech- nologies, as well as the development and growth of the industry, depend on the investments and actions of small companies will- ing to take linked technological and business risks. Small companies play an unusual role in the advanced display industry. in the early 1990s technological uncertainty, widely ex- pected but unrealized large markets, and government (Department of Defense) funding created a cadre of small relatively research; intensive companies in search of a valuable proprietary technologi- cal position. As evidenced by the high rate of partnerships with large companies, most small companies do not expect to be able to exploit such a proprietary position alone; the capital requirements of scale-up and manufacturing for large markets prevent most small companies from entering high-volume markets. In the ad- vanced display industry, small companies have been the reposi- tory of technical capability during the development of both core and complementary technologies and products. The cost of manufacturing facilities and the demands of process research and refinement suggest that these products will have to be pro- duced by larger companies or partnerships. While there are considerable differences between outdoor sporting goods anct implantable and surgical medical devices, some similarities stand out. in particular, user-inventors in both indus- tries play an important role in bringing new products to market. In both industries small companies and user-inventors have provided innovation and product diversity in small markets. Moreover, in both industries most of the new technology that becomes part of a new commercial product comes from outside the industry itself. The considerable differences are also evident. In medical device innovation, most user-inventors are highly technically trained, the "field testing" of products is often a formal clinical trial in a univer- sity hospital, and, historically, venture capital was widely avail
TECHNOLOGICAL START-UPS AND Sail INNOVATIVE COMPANIES 37 able for medical device start-ups. Further, medical devices are sold to physicians (and increasingly to health care organizations) rather than retailed clirectly to consumers and, of course, they are increas- ingly regulated by the FDA, a change that industry observers be- lieve may alter forever the way medical device innovation is done and, perhaps, eliminate small company participation. The out- door sporting goods industry and the implantable and surgical medical device industry illustrate the range of small markets that are served by small innovation-oriented companies that often pull fundamental technical advance from outside the industry into new applications. Among the industries addressed by this study, environmental testing services stands alone. The industry is, indeed, a technically oriented industry dominated by small companies, but it has a unique characteristic because demand for services is created by regulation and driven by enforcement. Barriers to entry into the industry are Tow but so are the opportunities for growth; econo- mies of scale, if they exist at all, are small and the opportunities for innovation are slight. However, environmental testing laborato- ries serve an important public function. Without good testing ser- vices, environmental regulation, and some aspects of environmen- tal protection, would not be possible. in this sense, environmental testing laboratories are small companies that serve the economy by providing an important technical service in small geographi- cally differentiated markets. CONCLUSIONS As the industries studied for this project show, small high-tech companies play a critical anct diverse role in creating new products and services, in developing industries, and in driving technologi- cal change and growth in the U.S. economy. The contribution of these companies is not adequately measured by their R&D expen- ditures, their employment, or their contribution to the national storehouse of published or patented technical know-how. Pioneer- ing new markets and providing innovation in small markets are often inherently risky activities and may yield economic returns that, for larger companies, are not commensurate with the risk. In consumer markets, substantial benefits of linked commercial and technological risk taking the rapid exploration of market po
38 RISK AND INNOVATION tentials, the clevelopment and refinement of new products and ser- vices, certain types of technical innovation, and product diversity- accrue directly and immediately to individual consumers and to the national economy. In intermediate or industrial markets, the willingness of small companies to shoulder technological and related market risks may enable other companies (both large and small) to pursue otherwise unavailable technological approaches to improving productivity or introducing new products. In small consumer and intermediate markets slowly growing markets with total annual demand in the tens of millions rather than hundreds of millions of dolIars small companies are often the only source of products or services ancl, therefore, responsible for all product diversity and for bring- ing innovations to the market. These market functions are an important part of industry and national economic development and, as such, are key in the economy's learning process about which technologies, at what price, will or will not satisfy demand. As a consequence of different types and levels of business risk, however, small companies and start-ups fulfill market develop- ment functions in different ways in different sectors of the economy. Generalizing on the industries addressed in this study, the specific contributions of small technically oriented companies in an industry depend on the following factors: · The size of markets, current and prospective, that make up the industry. · The degree of technical uncertainty in current or prospective markets. · The economies of scale and scope in production for a market. · The dominance (or lack thereof) of larger companies in both final and intermediate markets in the industry. It is important to emphasize the last point that many high- tech opportunities for small companies depend intimately on the current structure and operations of large companies in an industry. For example, economies of scale and scope in the aircraft industry exclude start-up companies from the business of bringing a new airframe to market. It is, however, the structure and operation of large airframe manufacturers that determine whether a small high
TECHNOLOGICAL START-UPS AND SMALL INNOVATIVE COMPANIES 39 tech company can supply new materials or advanced software to the industry. Further, there is often a bias in large companies against certain types of technical investments; although large companies and small companies may see the same technological needs and opportuni- ties, they perceive risk and make investment decisions differently. For example, large companies often do not address small markets on which they might make an acceptable level of return on invest- ment, because the absolute value of the reward is insufficient to be noticeable on the company's books or because it cannot be quanti- fied the return on investment is not initially known for "seminal" investments, and so larger companies do not pursue them. Incre- mental investment decisions by large companies are likely to focus first on improvements in existing businesses and second on enter- ing markets that have a good prospect of growth to a substantial size. From the perspective of a large company, this allows them to focus and to share with other companies some of the technological risk of improving their current business. Based on the industries addressed in this study, the committee concludes that small technically oriented companies often assume types of risk (and an amount of risk) that is not usually tolerated by large companies. in the United States both consumers and cornpa- nies often depend on small high-tech companies to explore the commercial application of technology in potential, emerging, and small markets. The principal economic function of small entre- preneurial high-tech companies is to probe, explore, and some- times develop the frontiers of the U.S. economy products, ser- vices, technologies, markets-in search of unrecognized or otherwise ignored opportunities for economic growth and devel- opment.
