A while ago I posted about what I called hypocrisy, in the form of conservatives' inability to imagine what it's like to be underprivileged or underrepresented unless someone in their immediate family brings it home to them. You might consider that a lack of empathy. Now there's a recent essay in The New Yorker called "The Case Against Empathy." Basically, empathy is horrible at math (read: cost-benefit analysis). Also, conservatives claim they're being empathetic too, just with the over-regulated small business owner instead of the polluted-on inner-city minority, so just claiming we need more empathy doesn't really solve many arguments. And today there's a short bit in The Atlantic about the same thing: military intervention to stop atrocities is really expensive, so likely less cost-effective than other ways to save lives.
I suppose it just boils down to your definition of empathy. If you think it's a blind emotion with no reason, sure, it's not that useful, but that's really only a straw man. The New Yorker piece talks about everyday gun crimes compared to massacres, and preventable diseases and starvation compared to Katrina or the Boston bombing, and says that Americans' over-fascination with the latter of those pairs, along with babies who fall down wells and white girls who get abducted in Aruba, is somehow proof that those who call for a "global empathic consciousness" are wrong. But that's exactly the same point that those who call for more empathy are making--we'd be better off if we spent less money on American babies who fell down wells than we spent on malnourished sub-Saharan African children, and we'd be better off if we sent fewer stuffed animals to the wealthy suburb of Newtown, Connecticut and more cash to something like Haitian orphanages. We're already pretty good at showing empathy for people who are similar to ourselves. It's people who are different that we need to work on.