What they say: Coyotes are free pest control.

What's true: They do eat rodents. But they crowd out foxes, which are more effective predators of rats.

But "free pest control" describes a secretive coyote hunting in a ravine. That is not the animal in dispute. The coyote that gets a file opened on it is the one eating out of a green bin.

That distinction shows up in the diet research. In a southern California study using both scat and stable isotope analysis, human food made up 22% of scats and an estimated 38% of diet by isotope analysis, with domestic cats and ornamental fruit also significant items — and consumption of anthropogenic items fell as urbanization fell. A northeastern-cities diet study found the site with the heaviest human footprint produced the greatest share of anthropogenic and urban commensal species in coyote diet. The more urban the coyote, the less of its diet is the rodent control being advertised.

And the animals in conflict are specifically the ones on that diet. In the Edmonton collaring study, coyotes visibly infested with mange used more developed areas, had larger home ranges, were more active during the day, and assimilated less protein than healthy coyotes — the authors' read being that human food is a low-quality but easily accessible option that increases reliance on anthropogenic resources and drives encounters with people. That is the same profile as the escalation markers in the Murray/Foucault/Patton paper: daytime activity in high-use areas, close approach, attractant exposure.

So the trade being offered is false. Targeted removal of identified problem individuals doesn't try to change population size, and the animals doing the rodent work are the ones nobody is filing 311 reports about. You keep the ecosystem service either way.

Sources:

Larson et al. 2020, PLOS ONE — https://doi.org/10.1371/journal.pone.0228881 Baseline diet of an urban carnivore, J. Urban Ecology 2020 — https://academic.oup.com/jue/article/6/1/juaa021/5979496 Murray et al. 2015, Proc. R. Soc. B — https://doi.org/10.1098/rspb.2015.0009