Right, the quote from the linked paper is:
> This is contrary to a suggestion that finding fewer insects on number plates in recent years might be attributed to increasing streamlining of vehicle aerodynamics over time. Our data show that newer vehicles sample more insects
than older vehicles, and we have observed pattern of fewer insects on number plates more recently than in the past in spite of this effect of vehicle age, which is assumed to be correlated with aerodynamics.
In fact the researchers posit a few more possiblities other than collapse of insect populations that could explain their data. For example they can't eliminate the possibility that there's some change in the management of roadside verges that specifically decreases the insect population in those environments, without reducing the population in other environments.
The challenge for all the explainations which involve something other than population change is that there's now an increasing nubmer of studies using different techniques which consistently show declining populations. For example [1] uses Malaise traps in nature reserves, which presumably wouldn't be affected by roadside management practices, and finds a ~75% decrese in flying insect biomass over 27 years, whereas [2] uses light trap data to look at moths specifically and finds a ~40% decline in biomass over 50 years (and ~50% in woodlands).
Whilst it's possible to find studies that show increases in insect numbers e.g. as a result of site-specific conservation work, there's a lot of data suggesting that in Europe at least numbers are in significant decline. As far as I can tell (and I'm really not an expert) we don't understand the reasons all that well, but habitat fragmentation and pesticide use are leading theories. We at least know that common pesticides (notably, but not exclusively, neonicotinoids) are toxic to bees, and will reduce populations even at sub-lethal doses (e.g. [3]). So changing our agricultral practices to eliminate these chemicals looks like a necessary, but perhaps not sufficient, part of the solution.
[1] https://journals.plos.org/plosone/article?id=10.1371/journal...
[2] https://resjournals.onlinelibrary.wiley.com/doi/10.1111/icad...
[3] https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.66...