Answer
An efficiency of 0.713 is the proportion between the sample sizes of the corresponding parametric test and the given nonparametric test, so that both tests have the same results:
$\frac{sample~size~parametric~test}{sample~size~nonparametric~test}=0.713$
Work Step by Step
That is, if a sample size of 1000 was obtained for the nonparametric test, then, if you want to achieve the same results, a sample size of 713 will be required for the corresponding parametric test.