Calculated risks and the costly status quo

Great circle route

The next time you travel by airplane, look out the window and see if you can count how many engines are attached to the wings. Chances are pretty good you will find only one on each side. This holds true even on routes with long stretches over water or harsh terrain that provide no suitable diversion sites in case of mechanical trouble.

With few exceptions, most jets in commercial service now come fitted with two engines, a notable change from the status quo in the middle of the 20^th century. In those earlier days of air travel the norm was to have four, and not because they provided the optimal ratio of power or efficiency. The main reason for this redundancy was the perceived unreliability of existing engines.

If there were only two to begin with, a blown piston or other mishap would leave just a single engine operating, a prospect too risky for regulators. This led manufacturers and their airline customers to converge on four engines as the standard. (For obvious reasons of symmetrical thrust an odd number wasn’t a popular choice, although some models featured a third engine embedded in the tail.)

What’s more, regulatory bodies like the Federal Aviation Administration in the U.S. mandated that aircraft couldn’t stray too far from possible landing sites, in case of emergencies requiring immediate help from the ground. This meant routes were carefully plotted not to take the shortest distance between origin and destination but to stay within range of potential diversion airports throughout the flight.1

Planes with four engines were given more leeway in how far they could be from emergency landing sites as multiple simultaneous breakdowns were improbable, so for the longest trips they were still king. Such policies enhanced safety but came at a price, as each deviation from the ideal path used up time and fuel. In certain cases this incremental cost of the rules made service between city pairs economically unviable.

With the advent of simpler and highly reliable jet turbines, regulators began relaxing the restrictions on how closely aircraft needed to remain to possible landing sites mid-route. Under the rubric of “ETOPS”, standing for Extended Operations,2 operators were able to prove through rigorous safety certification that in-flight failure was unlikely, and manageable should it occur, so tight constraints were not needed.

An original rule requiring planes to be within 60 minutes of an airport was relaxed, first to 90 minutes, then two or even three hours, until today there are some craft certified to fly over six hours away from any runway. The implications sound extreme at first—passengers on such planes trust they can remain airborne for six hours in the event of a technical problem that can’t be solved while in the skies.

Fortunately, these modified regulations have been standardized globally and steadily expanded without incident. Planes today are designed to be perfectly airworthy if all but one of their engines fail, and such issues are exceedingly rare.3 The prior approach has been discarded as unnecessarily conservative.

This is a valuable change for travelers who now have direct options between locations that would not be profitable for airlines to service were the more cautious routings still required. Airlines save on fuel, and with fewer expensive and complicated engines to service they have lower capital and maintenance costs. Manufacturers have in turn designed massive new powerplants that operate with record-breaking efficiency. New planes from Boeing and Airbus have globe-spanning reach and take for granted that extreme redundancy won’t be required.

Known unknowns

Policies like ETOPS might seem foolhardy—after all, why not add extra safety requirements just in case the worst happens? If they were costless that wouldn’t be a problem, but in the real world of aviation more backup adds more weight and drag. Circuitous routings cost time and fuel, which can be calculated with precision. Even still, these known costs may be preferred over the unknown cost of an emergency, since ambiguity is unsettling.

This reflects a broader psychological phenomenon known as risk aversion, which has been well-attested in the social sciences. Humans prefer to avoid losses and will incur real costs to mitigate them, even if these costs are greater than the expected loss. This effect works at the level of organizations, or even more complex systems like aviation regulation.4

When evaluating the merits of a new path the downsides can be obvious, since what you have to lose is generally clearer than what you have to gain. For instance, if a company launches a new product it may risk cannibalizing its existing profitable ones. The status quo will have managers who are quick to point out that fact, and those whose performance evaluations are contingent on it will provide especially sharp calculations of the risks of change. That consideration can overwhelm the fact that their market share is already under pressure and competitors are circling with their own offerings.

Building in every possible contingency as part of a strategy can end up producing something so encrusted with extraneous elements that agility is compromised. Alternatively, it may hew so closely to known and safe paths that it ends up losing the novelty that would make it compelling. If you can’t cut yourself loose from a certain strategy or mental model, your degrees of freedom become limited. In the process new paths are closed off, even though they might unlock a different way of operating. Sometimes caution is a crutch whose real costs are not adequately calculated. A better path might involve getting rid of the safety net.

When faced with ambiguity too often we choose the guaranteed loss, which might be greater than the as-yet unknown costs of taking the riskier path. The safe route may be comfortable, but it is costly.

---

Not receiving these articles automatically? 1/week, subscribe here.

---