I think it's simple: fat unhealthy men have lower sperm counts and America has way more fat unhealthy men over the last 40 years. Doesn't look like the meta analysis dealt with this.
While the US certainly has obesity epidemic, the results held across a number of countries in Europe and Oceania as well (and not in the African, Asian, and South American countries investigated; so the “Western” label is misleading, given that South America is as much part of the West as North America.)
I don't think obesity is the differentiating factor here.
It's not too much of a stretch to consider that there may be metabolic conditions in men linked to obesity risk that have the same kind of advantages in reduced-food-supply situations.
it is also depends on how much more energy one has to expend to haul that stored energy around.
The situation is probably too complicated for a crude model like this to really apply well, but it seems to me the rough sketch of it should be more like playing Yahtzee and rolling for at least one six with a subset of the dice instead of all the dice. (In that case the chance to succeed is the complement of the product of all chances to fail)
I think this is a very good and relevant question. What does it actually mean for fertility? Quality of the sperm rather than quantity is probably a much more important factor for that. (What a shame that's much more difficult to assay...)
But regardless of whether it leads to a marked impact on fertility, for an important biological function to drop like this in so large populations is a big blaring air-raid siren of a signal that someone needs to sort out why and fix it.