I find it doesn’t make sense to compare annual hours worked by employee.

Instead, only annual hours by person living in that region should be compared. Because otherwise, more part-time workers (meaning more working hours in total) dilute and decrease the average.