Variance inflation factor is calculated using each independent var as the dependent var, regressed against the rest of the vars. Does Stata actually run all of those regressions under the hood or is something quicker done?

Thanks,
Alex