The effect of information on household water and energy use

Abstract

Water and Energy Utilities are faced with growing demand at a time when supply expansion is increasingly costly, inconsistent and taxing on the environment. Given that supply expansion is limited, to meet future needs utilities need demand-side management policies to result in more reliable and consistent consumer responsiveness. Currently, most households do not have access to the level or type of information needed to respond to price signals in a reliable and effective way. Advanced information technology solutions exist and are being increasingly adopted, but we need to know more about how the informational setting affects decision-making, consumption levels and price responsiveness. This research analyzes the effect of information on household water and energy consumption, which is a decision-making environment characterized by uncertainty and imperfect information. This study also analyzes additional complexities stemming from infrequent billing, non-linear pricing structures, and combined utility bills, each of which may dampen price signals. I first develop a theoretical model of decision-making under uncertainty. I use this model to illustrate the effect of more frequent information, which eliminates uncertainty about past decisions, on remaining decisions within the billing period. The model emphasizes the role of risk preferences and the realization of the uncertain quantity. On average, risk averse consumers will increase consumption when uncertainty is reduced; risk seeking consumers will do the opposite. Introduction of a non-linear rate structure induces behavior that makes individuals appear as if they are risk averse or risk seeking, despite their actual risk preferences. This model highlights the importance of modeling multiple decisions within a billing period and accounting for a spectrum of risk preferences. In Chapter 3, I create a computerized laboratory experiment designed to generate data used to test some of the hypotheses formulated in the theoretical model presented in Chapter 2. Results from the experiment show that, on average, individuals consume more when provided with more frequent information that resolves uncertainty about past decisions made within a single billing period. This result is driven by the fact that the majority of participants are risk averse or risk neutral. Risk seeking participants instead reduce use when facing less uncertainty. Also as predicted by the theoretical model in Chapter 2, combining behavior driven by risk preferences with the presence of an increasing block rate structure results in behavior that looks like consumers are targeting the block boundary. This experiment shows that providing more information may not lead to reduced use without other incentives, goal-setting, or mechanisms designed to help individuals process the information. In Chapter 4, I empirically analyze a ten-year household-level panel data set of monthly utility bills. A single utility provides electricity, natural gas and water services to its customers and therefore bills through a single utility bill. I first show that price responsiveness varies by the number and combination of services subscribed to by a given household. Second, through a price salience model I show that households are more responsive to the price of water when the water portion of the total bill is greater. When multiple services are contained on a single bill, the salience of any individual price signal is dampened. This study confirms that households are inelastic though not unresponsive to water prices. In order to make pricing policies more effective, utilities need to acknowledge that households may be responding to total utility costs (i.e., may respond to a high utility bill by reducing electricity use despite the true driver of the high bill) and will need to find ways to make quantity and price information more salient to their customers. Chapter 5 concludes this dissertation by summarizing the contributions of the research and possible extensions for future work. By improving the informational environment surrounding household water and energy use, there will be great capacity for households to use water and energy more efficiently and ultimately make choices that reduce residential water/energy consumption and yield benefits for customers, utilities, and the environment.