<p>In academia, diversity in education and experience is valued highly, so it makes sense that programs would prefer that their undergraduates have at least some advanced education elsewhere. If a researcher only knows, say, Berkeley, their world is much more closed than a researcher who knows Berkeley and MIT. </p>
<p>Likewise, top programs generally do not like to hire their own newly minted PhDs. Again, it seems perverse, especially given that the top students of a top program are among the best of the best, but that’s the (usually unwritten) rule of academia. </p>
<p>Mollie tells me that MIT is the exception in all this, since it accepts its own undergraduates to its PhD programs and hires its own PhDs, but most institutions do not like to do this.</p>