There is a trick here. The Economist has equated risk with breaking the law. It might not be illegal to run a bank into the ground, however the Savings & Loans scandal of the 80's didn't result in bankers going to jail on trumped up charges. It is illegal to fix the LIBOR rate through collusion, and that has nothing to do with taking risks. Funny that the author cites clear evidence of criminal behavior and then goes on to argue that such evidence is difficult to come by.

Rather than spending their time looking for new ways to punish bankers, policymakers would do better to find ways to avoid having to bail out banks out in the first place.

This is a bizarre statement considering that so few bankers have been charged, and that charges require evidence of criminal activity, not simply taking financial risks. The author (the anonymous author in the column "The Economist Explains"), argues that it's not worth prosecuting bankers for criminal behavior, because their job requires taking risks. The author doesn't want you to consider that those risks should take place in a legal framework, or that bankers are currently at very low risk of prosecution for working outside of it.

Countries such as the United Arab Emirates that penalise bankruptcy and even threaten debtors with prison are not known for their vibrant start-up cultures.

How many countries are known for their vibrant start-up cultures, and which of these tend not to prosecute bankers for criminal activity? Are bankers in Russia often prosecuted for criminal activity? How vibrant is their 'startup culture'? A statement like this one about the UAE is meaningless, but the author uses it to make a point.