Conference Agenda

Households with sufficient financial means are adopting solar home systems (SHS) to mitigate the impacts of load shedding in Cape Town, leaving lower-income households that are unable to afford the high upfront costs to bear a disproportionate share of the impacts from unreliable electricity connections. In this paper, we investigate whether SHS adoption significantly affects household electricity consumption during load shedding, suggesting that SHS enhances resilience, benefits largely confined to higher-income households. We use panel data on household electricity transactions and load shedding from the City of Cape Town (CCT). In addition, we built our own deep learning model using the Mask2Former infrastructure to identify the locations and capacities of SHS within the city. Our fixed-effects model results indicate that households with SHS generally consume more electricity from the grid but reduce consumption during load shedding, suggesting they smooth their consumption with their backup generation. On the other hand, households without SHS experience no impact on consumption during outages, indicating any consumption that could have been lost is pushed either before or after to compensate. These results show that SHS are effective in improving a household's resilience to unreliable power, but that impacts are most likely concentrated among the wealthier population.