After a thunderstorm, I was disempowered for about 5 hours today. Certainly not the end of civilization, but perhaps its foreshadowing. A few moments ago, the power truck rolled down my alley, made some adjustment, and my house roared back to life. I have been re-empowered.
This raises the question, why are power lines still above ground?
Richard Layman sends me to this Electrical Industry discussion of the issue. My sense is they would be happy enough to put utilities underground so long as someone else pays. While underground utilities are less likely to fail due to storm, they may take longer to restore.

If electricity costs me about $0.10 an hour, ($2.40/ day, $876 year), then I would be willing to pay at least $0.10 to avoid an hour of blackout. In all likelihood, I would pay much more than that. In a typical year I am probably blacked out for 24 hours.
If converting to underground distribution cables for utilities costs $723,000 per mile (let’s round to $750K, there is a very wide range of suburban costs of new distribution construction according to the report), and there are about 100 houses per linear mile (a convenient guess, 10 houses per block * 10 blocks per mile (at uniform density, assuming square lots, this implies a density of 10,000 houses per square mile of residentially developed area or 23,000 persons per square mile, which seems high, but we are ignoring areas that don’t have houses as they don’t need residentially-oriented electricity wires), and the line can serve two row of houses (i.e. it runs in the alley) the cost is about $3750 per customer.
I would need to avoid 1562 days of blackout at $0.10 per hour to justify this on blackout avoidance. (In other words, ignoring discounting, if I can avoid 1 blackout day per year, it would take 1562 years to pay back). Obviously I am probably willing to pay more (reducing the payback time), I might even pay $100 per blackout day in extreme cases (maybe the cost of a hotel stay), but that still requires a 37.5 year payback, which is far more than most people would be willing to tolerate. Given the differences in reliability between above and below ground, undergrounding is not economically justified as retrofit for the purposes of continuous electricity unless power outages get much worse.

There are other advantages. Aesthetics for one. And I think this is important, though everyone will weight this themselves. One study in Australia suggests that underground networks increases house prices by 2.9 percent. For an average house price of at least $129,310 this would mean it is worth at least $3750. Now it pays for itself. A stated preference survey by one of the same authors also in Canberra estimates value of $6883 per house.
James Fallows discusses electric infrastructure reliability in the wake of the derecho back east.