It is commonly accepted that the existence of polar mesosphere summer echoes (PMSEs) depends on the presence of charged aerosols since these are comparatively heavy and reduce the diffusion of free electrons due to ambipolar forces. Simple microphysical modeling suggests that this diffusivity reduction is proportional to rA2 (rA = aerosol radius) but only if a significant amount of charges is bound on the aerosols such that NA∣ZA∣/ne > 1.2 (NA = number of aerosols, ZA = aerosol charge, ne = number of free electrons). The fact that the background electron profile frequently shows large depletions (biteouts) at PMSE altitudes is taken as a support for this idea since within biteouts a major fraction of free electrons is missing, i.e., bound on aerosols. In this paper, we show from in situ measurements of electron densities and from radar and lidar observations that PMSEs can also exist in regions where only a minor fraction of free electrons is bound on aerosols, i.e., with no biteout and with NA∣ZA∣/ne ≪ 1. We show strong experimental evidence that it is instead the product NA∣ZA∣rA2 that is crucial for the existence of PMSEs. For example, small aerosol charge can be compensated by large aerosol radius. We show that this product replicates the main features of PMSEs, in particular the mean altitude distribution and the altitude of PMSEs in the presence of noctilucent clouds (NLCs). We therefore take this product as a proxy for PMSE. The agreement between this proxy and the main characteristics of PMSEs implies that simple microphysical models do not satisfactorily describe PMSE physics and need to be improved. The proxy can easily be used in models of the upper atmosphere to better understand seasonal and geographical variations of PMSEs, for example, the long debated difference between Northernand Southern hemisphere PMSEs.