The Economist published a leader in its June 2, 2012 issue in support of hydrualic fracturing, a method for extracting otherwise difficult to recover natural gas. The editors addressed some of the risks associated with fracking and came to this conclusion (emphasis added):
But the risks from shale gas can be managed. Properly concreted well-shafts do not leak; regurgitants can be collected and made safe; preventing gas venting and flaring would limit methane emissions to acceptable levels; and the risk of tremors, which commonly occur as a result of conventional oil-and-gas activities, can be contained by careful monitoring. The IEA estimates that such measures would add 7% to the cost of the average shale-gas well. That is a small price to pay for environmental protection and the health of a promising industry.
The editors are right to note that there are engineering solutions to mitigate the risks associated with fracking. But they fail to address (1) the costs associated with failure in any of these measures and, more importantly, (2) the degree to which we can realistically expect these measures to be adequately undertaken and enforced.
The operative words in the quote are in bold, with an emphasis on “can”. Can reflects on capabilities, not actualities, which are far more important when discussing risks.
Yes, hungry dogs and newborn babies can be left alone together, but we expect the experiement to end very poorly for an unacceptable number of trials. People can be educated about the dangers of household fires, but we expect that people will still forget and burn down their houses. And they do.
The risks associated with fracking are externalities, and the measures necessary to mitigate them, properly or not, represent costs. So firms have little economic incentive to mitigate risks which do not impact extraction, and they have an economic incentive to skirt regulatory requirements which impose a cost burden. The extractive industries are exemplar of this behavior.  
It is entirely possible that a government apparatus could be set up to properly regulate the practice of fracking. An agency can do this well, but the libertarian-minded Economist would probably be more sensitive to the incentives of the bureaucracy. For better or for worse the overseer probably wouldn’t be the EPA, but rather a more ‘expert’ agency to oversee the frackers. An expert agency subject to or wholly captured by industry.
The risks of fracking can be mitigated, and the practice can be regulated to provide reasonably safe, clean energy. But let’s not kid ourselves that the power of that word extend to out expectations.