Electricity access is an important driver of economic development. Previous studies treat electrification as a binary outcome. In reality, in developing countries households with access face chronic supply interruptions, which can last up to 12 h a day. This is the first study to estimate the income differences in urban and rural non-farm enterprises in Indian households with different levels of electricity supply, using a subset of 8125 households in the India Human and Development Survey, a cross-sectional national sample of 41,554 households. I use multiple econometric approaches, including linear regression with an instrument variable and propensity-score matching with multiple treatment levels to represent supply availability. I find a robust income effect of access, and suggestive evidence of the effect of better supply availability. The aggregate income impact across existing NFEs in India of improving supply to 16 h a day could be on the order of 0.1 percent of GDP.