<p>The Pareto distribution and its generalizations are commonly used in modeling different phenomena in economy and natural sciences. In this paper we consider an array of i.i.d. random variables which have an asymmetric Pareto-type distribution and prove the strong exact law of large numbers for extremal statistics taken in each row in the case when the length of the rows tends to infinity, the case of fixed length of the rows is also considered. As a tool we prove a general exact law of large numbers for nonidentically distributed random variables with the left tails dominated in a sense by the right ones.</p>