Schilling:Outsourcing_book_temp.qxd 26/08/2009 10:41 Page 9
Industry Outlook
Plotting Renewables and Fossil Fuels on the Technology S-curve
a report by
Melissa A Schilling and Melissa Esmundo
Stern School of Business, New York University
Despite the well-known disadvantages of continued reliance on fossil can obscure the true relationship, since effort may not be constant
fuels, the vast majority of the world’s energy consumption still comes over time.
from fossil fuels and the majority of research and development (R&D)
investment in energy technologies is spent on further developing Cost Estimates of Renewable Energies versus Fossil Fuels
fossil fuel technologies. There are at least two major impediments to Table 1 shows a comparison of cost estimates for the various
greater use of and investment in renewable energies: renewable energy alternatives (with the exception of hydroelectric
power and biomass energy, for which yearly cost data were
• the current costs of renewable energy technologies; and unavailable) and the most important fossil fuel energy sources, as
• considerable uncertainty about which renewable energy well as a fossil fuel composite price created by the US Department of
technologies offer the most promise for yielding an efficient Energy based on the proportional use of each. For comparability, all
alternative to fossil fuels in the future. values have been converted to cents per kilowatt hour (kWh).
However, plotting the performance of renewable energy alternatives As shown, in 2005 fossil fuels were still less expensive on average
and fossil fuels against the R&D invested in them yields some than renewable energies, but wind power and geothermal power
enlightening – and important – surprises that ought to be taken into were getting close to fossil fuel prices. In particular, it is worth
consideration in future investment by both government and industry. noting that wind and geothermal energies were still on sharply
decreasing cost trajectories. Fossil fuels, however, were increasing in
Technology S-curves of Performance Improvement price due to the combination of having exhausted nearly all of the
If a technology’s performance is plotted against the amount of effort options for technological improvement and encountering the
and money invested in it, it will typically exhibit an s-shaped curve: diseconomies of scale inherent in reliance upon finite inventories of
slow initial improvement, then accelerated improvement, then natural resources.
diminishing improvement. This pattern occurs because in the early
stages of a technology the fundamentals are poorly understood, Research and Development
causing improvement to be slow. Great effort may be spent Investment in Energy Technologies
exploring different paths of improvement or in exploring different In order to foster the development of renewable energy alternatives,
drivers of the technology’s performance. As scientists or firms gain a the governments of the world have funded  R&D in a number of
better understanding of the technology, improvement begins to renewable energy technologies.
accelerate. Producers can focus their attention on those activities
that reap the greatest improvement per unit of effort, enabling The International Energy Agency, an agency initially established
performance to increase rapidly. At some point, however, during the oil crisis of 1973–1974 to co-ordinate measures in times
diminishing returns to effort usually begin to set in. As the of oil supply emergencies, tracks R&D spending on the energy
technology approaches its inherent limits, the cost of each technologies of its 26 member countries. Of these 26, data are most
incremental improvement increases and the s-curve flattens out. reliably available over the period 1974–2005 for nine countries:
Canada, Japan, The Netherlands, Norway, Spain, Sweden,
Plotting the s-curves of competing technologies can yield Switzerland, the UK and the US. Since only the data available from
considerable insight into their likely future performance and the these countries are used, these data are employed only to observe
relative payoff of investing in them. Although the shape of patterns and draw inferences; it cannot be stated conclusively that
technology s-curves is not set in stone (unexpected changes in the the patterns observed among these countries hold for all countries
market, component technologies or complementary technologies of the world. It is also important to note that because there are no
can shorten or extend the life-cycle of a technology), they are a
useful way of visualising how the competition between two or more
Melissa A Schilling is a Professor of Management at
technologies is playing out. For example, while a new technology
New York University Stern School of Business (NYU
might initially have inferior performance to existing technologies, an
Stern). Her primary research areas are technological
innovation and knowledge creation. She has been
s-curve analysis might indicate that the new technology has a far
published in various journals, including Academy
steeper performance curve and is poised to overtake existing Management Journal, Academy of Management Review
technologies, rendering them obsolete. A key factor in such an
and Organization Science. Before joining NYU Stern,
Professor Schilling taught at Boston University and
analysis is to plot performance (e.g. speed, capacity, cost) against
the University of Washington.
a measure of effort (e.g. R&D, man-hours invested) rather than
E: mschilli@stern.nyu.edu
time. Looking at the performance of a technology against time
© TOUCH BRIEFINGS 2009
9