Is that really the right statistic? Seems like the relevant one would be the number of maintainers whose maintenance work is company-funded. (Ex, I'd imagine it would be quite bad if most contributions were from company-funded developers but had to be upstreamed by non-company-funded volunteers.)