Photovoltaic Systems, the experience curve, and learning by doing : who is learning and what are they doing?

Abstract

The photovoltaics industry has been growing at extraordinary rates over the past ten years as a result of increased government support for the technology. Yet supporting the technology is expensive and there is uncertainty over the future rate of technological progress for photovoltaics. Experience curves have been an important part of the argument to justify continued government support of as well as private investment in solar technologies. In this thesis, I argue that a more sophisticated understanding of experience curves and their underlying mechanisms will be important if we are to preempt a renewable energy "bubble." I begin by providing a brief history of photovoltaic technology and policies that have supported end photovoltaic end markets. Next, I examine the use of experience curves for photovoltaic technology, finding numerous conceptual inconsistencies. Finally, based on an economic analysis of almost 55,000 photovoltaic systems in the United States, I attempt to disaggregate the dynamics of cost-reduction in photovoltaic systems, with a particular focus on the behavior of the systems integrators/installers. My analysis includes measures for experience, competition, and installer characteristics. This thesis contributes a better understanding of cost dynamics in photovoltaics and calls for a more sober and more sophisticated discussion about cost dynamics when considering renewable energy policy.