That’s great if you have the luxury of having sufficient IDer capacity to check each other’s work.
Many places and taxa don’t. When a large portion of observations in a given taxon/region have clearly never been looked at anyone knowledgeable at all, it doesn’t seem like adding a third or fourth agreeing ID to observations is the most effective use of IDers’ time, particularly when the agreeing IDs are being added to those few species within the taxon that are generally ID’d correctly even by non-experts. And yet there are plenty of experienced specialist IDers who seem to prefer to do this.