Why should you want _arguments_ about a question for which there has been actual research? "Superior work outcomes" is a fuzzy thing to measure, but when units of work are clearly countable and it's agreed what constitutes "done", there are multiple studies that have found that WFH employees were more productive (i.e. did more in the same period of time), and since the pandemic there have been multiple findings that WFH employees were putting in more work hours (i.e. previously dead commute time became productive). Admittedly, the kinds of work that support this kind of clear measurement are generally not product design and development, but more like handling inbound customer calls, and for employees with some experience. The gold standard I'm thinking of here is Bloom et al from I think 2013 which did an actual random controlled trial. In that call center case, literally handling more calls per hour was attributed mostly to having a quieter environment, but workers did also work longer.
I think the main challenges aren't around "superior work outcomes" at the individual level, but stuff like training junior team members through interactions with their more senior colleagues. This is harder to measure, often isn't someone's "main" responsibility, and so individuals who appear to be succeeding at their own work can be less impactful on these secondary and indirect benefits.
All of which is to say, I think the personal fulfillment vs company benefit framing can be a misleading one. _which_ company benefits are most important? Does everyone in the company agree on that prioritization? Have you tried actually _measuring_ those, rather than just seeking "arguments"?