UC Berkeley

Net metering has become a widespread policy mechanism in the U.S. for supporting customer adoption of distributed photovoltaics (PV), allowing customers with PV systems to reduce their electric bills by offsetting their consumption with PV generation, independent of the timing of the generation relative to consumption. Although net metering is one of the principal drivers for the residential PV market in the U.S., the academic literature on this policy has been sparse and this dissertation contributes to this emerging body of literature.

This dissertation explores the linkages between the availability of net metering, wholesale electricity market conditions, retail rates, and the residential bill savings from behind-the-meter PV systems. First, I examine the value of the bill savings that customers receive under net metering and alternatives to net metering, and the associated role of retail rate design, based on current rates and a sample of approximately two hundred residential customers of California's two largest electric utilities. I find that the bill savings per kWh of PV electricity generated varies greatly, largely attributable to the increasing block structure of the California utilities' residential retail rates. I also find that net metering provides significantly greater bill savings than alternative compensation mechanisms based on avoided costs. However, retail electricity rates may shift as wholesale electricity market conditions change.

I then investigate a potential change in market conditions - increased solar PV penetrations - on wholesale prices in the short-term based on the merit-order effect. This demonstrates the potential price effects of changes in market conditions, but also points to a number of methodological shortcomings of this method, motivating my usage of a long-term capacity investment and economic dispatch model to examine wholesale price effects of various wholesale market scenarios in the subsequent analysis. By developing three types of retail rates (a flat rate, a time-of-use rate, and real-time pricing) from these wholesale price profiles, I examine bill savings from PV generation for the ten wholesale market scenarios under net metering and an alternative to net metering where hourly excess PV generation is compensated at the wholesale price. Most generally, I challenge the common assertion that PV compensation is likely to stay constant (or rise) due to constant (or rising) retail rates, and find that future electricity market scenarios can drive substantial changes in residential retail rates and that these changes, in concert with variations in retail rate structures and PV compensation mechanisms, interact to place substantial uncertainty on the future value of bill savings from residential PV.