Before presenting our package of recommendations, we offer here a broader view of how we expect PV technology to develop, and what general approach we believe will allow that development.

First devised at Bell Labs in the mid-1950s, photovoltaic technology has become commonplace for some comparatively narrow industrial applications (e.g., providing reliable power for some remote telecommunications facilities). Yet, much of the industry remains immature, and experts disagree on how best to develop it. On the one hand, the existing electric system may prove able to absorb PVs as a conservative innovation, allowing the technology to exist in comparative harmony with current institutions, technologies, and measures of value (such as the cost of energy). Indeed, PVs may help sustain and preserve the existing system. On the other hand, PVs may represent a radical, disruptive innovation, able to thrive only in a very different environment and possessing attributes not generally incorporated in established measures of value (such as adaptability to customer-located installation). On the basis of our investigation, we incline to the latter view: that PVs challenge the technological status quo, and can best be exploited by new institutions providing new kinds of value. To appreciate this point, consider the divergent ways in which conventional generating technologies and PV achieve cost reductions.

Nuclear and fossil-fueled power plants represent a constructed technology: unique, centralized projects connected to users through complex transmission and distribution networks, requiring several years and massive amounts of capital to complete. These technologies benefit from economies of scale. That is, up to capacities of roughly a gigawatt, bigger plants produce cheaper power. For most of this century, electric utilities delivered ever-cheaper electricity by building larger and larger facilities.

Are Some Innovations More Innovative? Historian Thomas Hughes, who has written extensively on the electric power sector, distinguishes between conservative innovations, which tend to preserve large technological systems, and radical ones, which spark the construction of new systems. In a similar vein, business analyst Clayton Christensen describes sustaining innovations, which improve product performance according to traditional measures of value, and disruptive ones, which perform poorly according to conventional measures — but outperform established technology in ways valued by emerging markets. See Thomas Hughes, “The Evolution of Large Tech-nological Systems,” in Wiebe Bijker et al., The Social Construction of Technological Systems: New Directions in the Sociology and History of Technology (Cambridge, England: Cambridge University Press, 1987), and Clayton Christensen, The Innovator’s Dilemma: When New Technologies Cause Great Firms to Fail (Boston, MA: Harvard Business School Press, 1997). Might renewable energy technologies qualify as radical or disruptive technologies? See Richard Hirsh and Adam Serchuk, “Momentum Shifts in the American Electric System: Catastrophic Change or No Change at All?” Technology and Culture 37 (April 1996), pp. 280-311, and Adam Serchuk and Richard Hirsh, “Condemned to Repeat? IOUs, History and Green Markets,” The Electricity Journal 11 (March 1998), pp. 76-86.

Photovoltaic technology is more like consumer electronics than a conventional power plant. Along with other renewable energy technologies such as wind turbines and solar water heaters, as well as non-renewable energy technologies such as gas microturbines and fuel cells, and even energy efficiency and energy storage devices, PVs are not constructed but manufactured. Rather than scale economies, PVs offer economies of mass production: the more units manufactured, the less each one costs. Historically, the cost of PVs has dropped 18% for each doubling of cumulative production.²

Viewed this way, PVs seem to face a typical chicken-and egg problem: higher production will drive lower costs, but selling enough PVs may also require lower costs. Yet the problem is neither so straightforward nor so intractable. PV technology is not simply a manufactured substitute for constructed power plants; it offers a different set of values that canny retailers can exploit. Unlike, say, a nuclear reactor, PV technology lends itself to small, distributed uses, close to where people consume electricity. Under the right conditions, customer-located PV units can forestall the need to upgrade an overburdened power line; more often, a PV facility in a remote location makes unnecessary any power line at all. PV systems can provide high-quality, reliable power for users who cannot afford an outage — say, to run a computer network. And, of course, “green” customers may appreciate freedom from a well-known list of environmental impacts associated with conventional energy sources: toxic air emissions, land and water degradation from mining, production of radioactive waste, and the release of greenhouse gases, among others.

Thus the secret to expanding markets for PVs seems to include both the exploitation of high-value markets and subsequent volume-driven cost reductions. To highlight the distinctions, consider the two roughly defined pathways historically put forward for PV development:

• The power plant path relies on large, grid-connected photovoltaic power plants as a conservative innovation within the existing electric system. Besides megawatt-scale facilities, this path integrates PVs into utility operations as substation and end-of-line upgrades. Such facilities may capture economies of scale, although much more modest ones than conventional generating technologies; they appeal chiefly because of the stimulus they deliver to manufacturing capacity. Yet since early PV power plants will be unable at first to compete economically in the wholesale power market with conventional technologies, this path probably requires large amounts of public funding. On the other hand, it requires less business and technical infrastructure, since it does not rely on retail sales and since utility professionals would probably construct the installations. The “power plant path” also requires little consumer education, except where necessary to endorse government support.

• The product path, an alternative approach, builds production volume through developing affordable, value-based products and expanding markets. In this scenario, PVs compete at the retail level, and take advantage of characteristics other than the cost of the electricity generated.

  We endorse the "product path."

  These include the reliability of PV systems, the technology’s adaptability to grid-independent deployment, the possibility of avoiding upgrades for transmission and distribution lines, consumers’ interest in environmental products, and so on. This path limits government involvement to increasing the diffusion rate of consumer products through setting market rules, making strategic purchases, and other innovative support. However, it requires the emergence of a sophisticated retail infrastructure, as well as the availability of consumer finance mechanisms and education. Above all, it requires that PVs be integrated into existing business systems, such as the home-building sector.

We endorse the “product path.” It will take longer than the “power plant path,” and it will require arduous coordination of numerous groups. Yet we find that it makes better use of PVs’ nature as a manufactured technology and does not try to force PVs into an inappropriate technological framework. We also find that the product path resonates with accelerat-ing regulatory and market shifts in the electric system. Most utilities, constrained by lower profit margins, have retreated from research and development. In contrast, restructuring has refocused the marketplace on applications over kilowatthours, on value over price, and on customers over technocracy — an ideal setting for the development of PV products. The product path has become an increasingly accepted theme at recent conferences on the state of the PV industry.³ In sum, it strives ultimately for the volumes that could conceivably allow PVs to compete at the wholesale level on price, but it does so through painstakingly building markets for retail products in which PVs can compete on the basis of value.

In sum, several key points emerge from the research described here:

• An important element of market expansion will be the progressive lowering of prices through economies of mass production.

• PV manufacturers can best achieve these economies through establishing self-sufficient markets for high-value products that exploit PVs’ characteristics, and that people want to buy.

• Effective market expansion requires coordinated movement by a variety of actors, including the federal and local and state governments, the PV industry and its trade association, the financial sector, non-profit environmental advocates, foundations, and others.

• Public policy must seek to enable consumer markets for PVs, for example by establishing fair market rules, rather than merely by subsidizing purchases.

One significant aspect of this investigation is notable in its absence. Analysts in the early 1990s looked hopefully toward the nation’s electric utilities to bring PV technology into common use. Through organizations such as the Utility Photovoltaic Group, government programs allowed some American electric companies to install notable numbers of PV systems. And although we decline to endorse the “power plant path” in this study, there might nevertheless conceivably be a role for electric utilities in the “product path.” On the contrary, our research team manifests decreasing faith in traditional electric utilities’ ability — or willingness — to adapt to the technological and market changes represented by PV power, particularly in light of the utility sector’s cautious response to restructuring. Indeed, today’s utilities too often hinder PV development, just as they did a decade ago. For this reason, our authors look largely to new institutions to deliver PVs to future consumers. Whether the utilities’ unregulated subsidiaries can fulfill this function as effectively as completely unaffiliated entities remains an open question, as does the issue of whether regulated utilities may necessarily play a larger role in marketing PVs in developing countries with immature consumer infrastructures. And, indeed, the utility companies of the next decade may be much more innovative institutions. On the whole, however, our team manifests appreciable skepticism toward the utility sector.

Table of Contents Previous section Next section