If I understand correctly, the extension of the argument (in light of your objection) is this: assuming two equally sized and skilled pools of employees, Compan

If I understand correctly, the extension of the argument (in light of your objection) is this: assuming two equally sized and skilled pools of employees, Company Diversity will spend less on salaries than Company Bigot - meaning those funds are able to be spent elsewhere, giving Company Diversity a competitive advantage in the market. Eventually Company Bigot goes out of business (or lowers their salaries). Of course we assume, in this thought experiment, that all the other things Company Bigot could do to become more competitive (invest in more employee training, better infrastructure, etc) are also available to Company Diversity. This isn't always the case. Maybe Company Diversity's hiring practices have alienated it from potential suppliers or clients who prefer Company Bigot's views. Specifically I'm thinking of Japan, where (apparently; source: this site) the cultural emphasis on the salaryman paradigm means that hiring from the significant pool of contract labor can hurt your company image as far as working with the established corporations is concerned.