I left the game in 2015, but even then third generation sequencing could close bacterial genomes for < $1k. Long reads make it trivial to detect all sorts of stuff (sample mixup, contamination, phage activity, in vitro evolution) that goes unnoticed with short reads. Since the conclusion of large scale draft genome sequencing papers always seems to be 'we need a bigger sample size' anyway, I feel the money would be better spent on studying individual strains of bacteria.