Bad developers are the gift that keeps on giving. Everything looks great, you ship, it mostly works, and then you spend four years struggling to build anything on top of what was written, squashing data-loss bugs that take weeks to track down and can never have their root causes fixed, etc.
Companies that end up in the latter situation are basically zombies. They're already effectively dead, but nobody knows this is the case for years, pouring time, effort, and money into a bottomless sinkhole.
Perfect coders can also serve as really good mentors on teams, and catch serious issues everyone else would have missed.
People who are Smart but don’t Get Things Done often have PhDs and work in big companies where nobody listens to them because they are completely impractical. They would rather mull over something academic about a problem rather than ship on time.
If you're a startup trying to find product-market fit before you run out of funding, better to ship a few bugs every week (in features that have an 80% chance of not existing or being rewritten anyway within a few years) than nothing at all for months.