The disproportion isn't in Social Security benefits consumption. The disproportion is in lifespan, and the difference in benefits consumption is a side effect thereof, because of the way Social Security is defined to work.
I'm not sure that it actually helps anything to cast it in the cart-before-horse way the article does.