A common concern with efforts to directly help some small businesses to grow is that their growth comes at the expense of their unassisted competitors. Mckenzie and Puerto (forthcoming) test this possibility using a two-stage randomized experiment in Kenya which randomizes business training at the market level, and then within markets to selected businesses. Three years after training, the treated businesses are selling more, earn higher profits, and their owners have higher well-being. Point estimates of the spillovers on the competing businesses are small and not statistically significant, and the markets as a whole have grown in terms of sales volume.