Assuming the data is correct (a big assumption), there are a few potential drivers for this.

First, you're likely dealing with a mixture of fields. For instance, I believe you're required to get a master's degree to teach in a lot of local jurisdictions, but teaching doesn't pay well. If the population of masters degree holders includes a lot of teachers, that would bring the average down. That's going to be true for a lot of fields with well educated people working low paying jobs in the public sector or at nonprofits. In contrast, many high paying jobs in fields like computer science or finance only require bachelor's degrees.

Second, people sometimes stay in school because they can't find a job immediately after undergrad. Around 2008-2012 or so, that was fairly common as a tactic to wait out the recession. It's possible that the candidates that got jobs immediately out of school interviewed better than their counterparts that stayed in school, and it's also possible that they're getting paid more now as a direct result of that.

Third, masters degrees mean less job experience. If you look at two identical 30 year olds, the person with the masters degree will have less real world experience because they were in school longer, so that may imply a lower salary.

If you control for the various factors above, you might get the more typical relationship with more education implying more pay, but it's possible that we have multiple confounding factors at play.