Scale bias is the extent to which technical change increases the productivity of large relative to small firms. I show that this dimension of technical change is important for inequality. To illustrate the mechanism, I develop a tractable frame- work where people choose to work for wages or earn profits as entrepreneurs and where entrepreneurs choose from a set of available production technologies that differ in their fixed and marginal cost. Large-scale-biased technical change lowers entrepreneurship rates and increases top income inequality, primarily by concentrating business income. Small-scale-biased technical change does the opposite. I show the empirical relevance of scale bias by identifying the causal effects of adoption of two general purpose technologies that vary in scale bias, but are otherwise similar: steam engines (large-scale-biased) and electric motors (small-scale-biased). Using newly collected data from the United States and the Netherlands and a range of identification strategies, I show that these two technologies had opposite effects on firm sizes and inequality. Steam engines increased firm sizes and inequality, while electric mo- tors decreased both. Consistent with scale bias (rather than skill bias), I find that adopting entrepreneurs were the main drivers of inequality increases after steam engine adoption.