No doubt if class sizes were reduced but everything else was held constant, the effect size would be minimal. But, wouldn't smaller class sizes allow for more time and effort to be spent on the things that do make a difference?

Yes, it is so obvious that all things being equal reducing class size should make a difference; but the operative word is ‘should.’ Given we have spent so much reducing class sizes why then has it not (operative phrase ‘has it not’)? So, to date, why has it not? Reducing class sizes (e.g. from 25-15 or 40-30) does have a small positive impact (d=0.21) on student achievement but nowhere near as much as what many people assume. For smaller class size to yield higher effects, the type of instruction needs to be re-conceptualized to ensure the needs of all students are met within whatever the class size. Teachers need to focus on strategies that are maximized in smaller or larger groups and apply these respectively” (John Hattie, personal communication, March 24, 2018).