I believe I've seen quite a few studies that shows that the longer workdays the fewer work produced per unit of time. I can't seem to find them right now so take it with a grain of salt, but it sounds like longer and fewer workdays would not be desirable from your company's point of view in that case.
What I personally would like to see it how applicable these results are between jobs. The jobs in question here are quite unskilled such as caring for the elderly, and are physically demanding. What works for them might not work for a programmer for example.
To your first point, I think a common misconception with this as developers and skilled knowledge workers is that we only look at our personal productivity. Yes, an individual programmer's productivity may be decreased by working 50 hours in a week, but if the company's productivity overall is increased, you'd be hard-pressed to make an argument against it. This is assuming that extended periods of 50 hours weekly (including all planning and meetings, not just programming) is below the level that would induce burnout in most, or that that's factored into whether it's "better" for the company or not.
The point here is that if you have 50 hours of developer time, you will get the most output if you spread it over more days than one long stretch. Therefore the 4 day workweek may be a tougher sell than the 6 hour workday.
What I personally would like to see it how applicable these results are between jobs. The jobs in question here are quite unskilled such as caring for the elderly, and are physically demanding. What works for them might not work for a programmer for example.