This would not really be a solution. All it would result in is more people taking work home, because job evaluations would still be based on how much progress one made on the primary project. To get ahead and get that next promotion, many people would still feel compelled to work extra (unreported) hours. I suspect a madate such as what you propose would simply push the long work-days into being unreported.
This is basically what happens in graduate school (in the US, at least). I just finished a PhD program; as part of our contrats students agreed to only work 20 hours per week. There is no way one could finish a dissertation working on 20 hours per week, so everyone worked longer hours. Sure, you could complain about how many hours you worked (and some people did), but that did not change the fact that only 10–20% of Astronomy PhDs get faculty jobs, so if you want that faculty job, you need to put in the hours to do great research, regardless of the mandated 20 hour limit.