Yuval Millo has an interesting post, with regard to Nassim Taleb’s argument about the rarity of events, compared to the effects they have on an organization. The discussion is about risk management, but I think there’s something wrong about it.
The argument made is really about how to mitigate rare-but-devastating events. As Taleb notes:
We summarize: If small probability events carry large impacts, and (at the same time) these small probability events are more difficult to compute from past data itself, then: our empirical knowledge about the potential contribution – or role – of rare events (probability × consequence) is inversely proportional to their impact.
Yuval’s response to this is two-fold: first, to suggest that the is a lack of independence between event and preparation for it. Organizations encountering (rare) events close the loophole behind them, making those rare events, well, even more rare. And second, he suggests that there is more about the shift from ‘hot-cognitive’ to ‘cool-routine’ practices that accounts for relative impact. In other words, organizations are able to mitigate common crises not because the crises are not devastating, but because the organization has an infrastructure in place to cope with it.
The question for me is in this vein, but made with a slightly different emphasis: to what extent is something an infrastructural risk versus a content risk. An infrastructural risk is revolution, while a content risk is regime change. The former calls into question a political system itself, while the latter is a potentially dramatic change of direction within the established infrastructure.
A more concrete example, less dramatic than political revolution. In 2005, for those who remember back that far, there was a fire in the NYC subway. Fires in the subway are actually not 100-year flood rare; they happen not often but not infrequently. This one happened near a particularly critical switching station. Note that the initial assessment was that the A train would be out of commission indefinitely – and that’s a very long time. So here, the crisis migrated from a content crisis (a piece of track, a car, some passengers) to an infrastructural crisis (a switching station). The point is that an infrastructural event might be completely rare, or it might be relatively routine. Yuval is probably correct that the MTA can now be aware of these crises and take steps to eliminate future ones. But the point is that these crises could happen rarely, and get fixed (or screw the organization), or they could have all the time and get fixed (or screw the organization).
The point is that it’s worth distinguishing ‘rare’ risk from ‘central’ risk – imagining a relatively common occurrence that breaks, or challenges, your infrastructure would be the best kind of risk mitigation, instead of worrying about the 100-year flood. They are not the same thing.