Bryan Bollinger

Peer Effects in the Diffusion of Solar Photovoltaic Panels

Social interaction (peer) effects are recognized as a potentially important factor in the diffusion of new products. In the case of environmentally friendly goods or technologies, both marketers and policy makers are interested in the presence of causal peer effects as social spillovers can be used to expedite adoption. We provide a methodology for the simple, straightforward identification of peer effects with sufficiently rich data, avoiding the biases that occur with traditional fixed effects estimation when using the past installed base of consumers in the reference group. We study the diffusion of solar photovoltaic panels in California and find that at the average number of owner-occupied homes in a zip code, an additional installation increases the probability of an adoption in the zip code by 0.78 percentage points. Our results provide valuable guidance to marketers designing strategies to increase referrals and reduce customer acquisition costs. They also provide insights into the diffusion process of environmentally friendly technologies.

Learning-by-Doing in Solar Photovoltaic Installations

The solar photovoltaic (PV) industry in the United States has been the recipient of billions of dollars of subsidies at the federal and state level, often motivated by environmental externalities and dynamic spillovers from learning-by-doing in the installation of the technology. This paper investigates cost reductions due to learning-by doing (LBD) using comprehensive data on all solar PV installations in California from 2002 to 2012. We develop a model of installer firm pricing behavior that allows for economies of scale, market power, and dynamic pricing to quantify both appropriable and non-appropriable LBD. We find strong evidence for both, suggesting a role for solar PV subsidies to improve economic efficiency by addressing a non-appropriable LBD positive externality. However, our results suggest that the California Solar Initiative cannot be justified by non-appropriable LBD alone.

Risk Transfer versus Cost Reduction on Two-Sided Microfinance Platforms

Microfinance can be an important tool for fighting global poverty by increasing access to loans and possibly lowering interest rates through microlending. However, the dominant mechanism used by online microfinance platforms, in which intermediaries administer loans, has profound implications for borrowers. Using an analytical model of microlending with intermediaries who disburse and service loans, we demonstrate that profit-maximizing intermediaries have an incentive to increase interest rates because much of the default risk is transferred to lenders. Borrower and lender interest rate elasticities can serve as disciplining mechanisms to mitigate this interest rate increase. Using data from Kiva.org, we find that interest rates do not affect lender decisions, which removes one of these disciplining mechanisms. Interest rates are high, around 38% on Kiva. In contrast, on an alternative microfinance platform that does not use intermediaries, Zidisha, interest rates are only around 10%, highlighting the dramatic impact of intermediaries on interest rates. We propose an alternative loan payback mechanism that still allows microfinance platforms to use intermediaries, while removing the incentive to increase interest rates due to the transfer of risk to lenders.

Information vs Automation and Implications for Dynamic Pricing

Essential resources like electricity and water can experience rapidly changing demand or supply while the other side of the market is unchanged. Short-run price variation could efficiently allocate resources at these critical times, but only if consumers exhibit short-run demand elasticity. The question for firms in these markets has always been how to enable this response. Randomized control trials are increasingly used to test dynamic pricing and technologies that can assist in response by providing information and/or automated response. But, the trials typically do not randomize short-run prices. This paper illustrates how demand from a randomly assigned control group can be used to test the effectiveness of different technologies in increasing short-term price elasticity. To do so, we use a non-parametric control function approach that eliminates the bias inherent in estimating short-term price response using only household random assignment. We find that only automation technology leads to the short-term price elasticity needed to justify real-time pricing.

Structural Analysis of Multi-Channel Demand

In this paper, we propose a structural framework to study multi-channel demand. Our model explains a comprehensive set of demand outcomes as a function of prices and retail store proximity, including the frequency with which consumers shop, how much they spend per purchase occasion, whether they buy from the online (web) or retail channel, and how they allocate expenditures among multiple product categories. We allow channels to convey different amounts of information about product categories, which in turn affects a consumers expected utility from purchasing in a particular channel. For example, physical inspection of goods in the retail channel can provide information about product fit and feel that is difficult to assess in the online channel. Another distinguishing feature of our model is that shopping trip expenditures are endogenously determined in the first phase of a multi-stage budgeting process, where consumers allocate their income by trading off utility for the outside option and the expected utility from optimal channel and category expenditure choices with the focal brand. A key methodological contribution of the paper is to advance a highly efficient algorithm to compute optimal expenditures, which facilitates joint estimation of the model parameters by maximum simulated likelihood.We estimate the model using the purchase histories of approximately 10,000 randomly selected customers from a firm that uses both online and retail channels to sell directly to consumers. The firm doubled its retail footprint over our two year observation window, providing a rich source of customer-specific variation in retail store proximity that we leverage to identify the demand effects of interest. We find evidence of channel complementarity through increased overall shopping frequency as the distance to retail outlets decreases, accompanied by increased substitution from online to retail formats. Our estimates imply a 10% reduction in retail store distance increases existing customer annual revenues by 0.53%, by increasing retail revenues 1.96% and decreasing online revenues by 1.43%. In a series of counterfactual experiments, we demonstrate how our model can be used as a decision tool for managers to identify promising locations for new physical stores and to explore channel-based price discrimination policies.