That's an interesting article, thank you for finding it. It raises several interesting questions, though I haven't read it closely enough to know if they get answered. One is, is photo-erosion the reason that the distribution of stellar masses (called the "initial mass function", IMF) turns over at around a half a solar mass, so we end up with fewer low-mass red dwarfs than it looked like we might get from the distribution around one solar mass. One imagines the presence of OB stars in the same cloud could start eroding away those lower mass stars, which the paper says is an inefficient way to make brown dwarfs but it might be a good way to cap how many you get, and how many red dwarfs too.
A second interesting question stems from the fact that the paper says photo-erosion would be an even worse problem for forming low-mass stars if they did not already have a good start on their formation (which creates a deep potential well for hanging onto their gas) by the time the OB stars form. Since we normally imagine that OB stars form much faster than lower mass stars, this suggests either that the OB star formation is more spread out over the cloud's star-forming process, or else happens only at the beginning and then there are not so many OB stars later on. For the first case, maybe many of the OB stars are being triggered by the supernovae of an earlier generation of OB stars. (It is commonly suggested that triggered formation like that happens, but this is saying that it must keep happening for quite a long time (millions of years) into the life of the star-forming cloud, as that is the only way such low-mass stars could get such a head start on their own formation.)
The second possibility, it would seem, is that triggered formation strongly favors low-mass stars, so there aren't as many OB stars around at that stage to do the photo-erosion. The bottom line would seem to be, you can't have lots of OB stars around at the same time the low-mass stars are starting to form, so you either have to spread out the OB star formation over a long time (so there are not as many at any given time), or you have to make them all form together at the start (so there are not so many later on as the lower-mass stars are forming). It sounds like the paper isn't really able to decide that, but they are able to put important limits onto how many OB stars there can be whenever the low-mass stars do succeed in forming.
Indeed this seems like an observational question to decide-- can't we just look at OB associations and look for evidence that either low-mass stars are having a hard time getting started, but the ones that are there seem older than the OB stars?
