In a perfect world, sure. But in the actual world that we live in, reducing false convictions to zero is not realistic. So we have to look for areas where investing our resources in fixing a problem will have a comparatively large beneficial impact. If it's 25/100, there's a lot to be gained there. If it's 25/100,000, there isn't.
How many innocents convicted, due to institutionalized, consequence-free perjury, is too many?
If you think 25 out of 100,000 isn't "that pervasive a problem", thank your lucky stars you're not one of those 25.