I think you're wrong. I think companies are completely correct in assessing the cost of a bad hire to be much greater than the loss of a good hire, assuming you're getting enough good hires to fill the positions available.
Now, this assumes that you're not missing out on great hires; in other words, it assumes that the good hires you miss are at the bottom of your hiring range. Your point might be that this can't be guaranteed, and while that may be true, it's unlikely that any tweaks to a given hiring process are going to bring them in either.