I feel it.

Cohen's d=-0.132 is considered a very small effect size. In other words: The effect is statistically reliable, significant, and disagreements between studies are very low, but not that big.

>The approximately 2-point decline in IQ in adolescent-onset frequent cannabis users is not to be clinically significant

In a population level this might be part of the explanation, but we would have to know the correlation between prevalence of use of cannabis and IQ decline in the population. In developed nations there is trend where IQ drops 0.2 points per year (2-points per decade) since mid 1990s. In developing countries IQ increases 3-points per decade as Flynn-Effect still works there.

Rastas say that "cannabis" isn't for stupid people, which is an observational conclusion. There are other sometimes profound and imediate negative effects cannabis consumption