Have you noticed that international flights have become more crowded lately?
The airline metric that describes the extent of this crowding is the seat factor, defined as the percentage of seats on the plane that are occupied. For all international flights operated into and out of Australia, the seat factor has grown from just 66 per cent in 1991 to 77 per cent in 2012.
The rapid growth in seat factors and congestion on planes has generated at least two major problems for airlines.
The first is that jamming more people onto the plane generates a less desirable travel experience. Waiting time for entry onto and exit from the aircraft, toilets and delivery of meals all increase when seat factors are high. There is also more noise in the cabin and more congestion in the common areas around exits, toilets and at the back of the plane.
The second and more serious problem is that despite the growth in seat factors, international airlines continue to experience below-target returns. This is a conundrum that has baffled airline executives and other aviation analysts for years because they believe that high seat factors reflect strong demand, which is correlated with favourable returns.
The flaw in their logic can be evidenced through four considerations – the breakeven seat factor, the failure of sophisticated yield management systems during economic downturns, preoccupation with market share, and the misalignment of the objectives of airline departments that decide on capacity and yield management.
The breakeven seat factor is a tool used by airline management to determine the number of seat sales that are needed to reach a profit target. If calculated properly it's an exceptionally useful tool.
In fact, it's not only a useful tool for airline businesses but other businesses that must decide on how much of their fixed capacity to utilise. How many rooms should be occupied at a hotel? How many seats should be filled at a cinema? Should football stadiums and rock concerts target a sell-out? And should restaurateurs try to fill all their tables?
The biggest problem airlines encounter when using breakeven seat factors is in the context of an increase in unit variable costs. Breakeven modelling tells them to raise the seat factor in response to higher cost, which is precisely what they have done for more than a decade. What they should be doing to recover higher costs, however, is the opposite – increase the price and reduce the seat factor.
If the seat factor falls too far in response to higher average fares, which is often the case when the economy is weak and consumers are shopping around for bargains, the airline should reduce capacity. To do this, the department responsible for managing yields should talk to the department responsible for managing capacity and together hatch a plan.
Often this is difficult to do because the objective of the capacity (or network planning) department is not aligned with that of yield management. The capacity people are trying to target a share of the market, maximise network coverage and offer high frequency at the peak at major airports – or all of the above.
Targeting a share of capacity or seats in the market, however, makes it almost impossible for the yield management area to optimally alter prices and seat factors in response to adverse movements in cost and underlying demand.
To illustrate, suppose that an airline's competitors grow their capacity at twice the long-run average pace (about 5 per cent) in an environment in which costs are rising rapidly. The capacity department of the airline says that the airline must match this capacity increase to preserve its market share.
In a world in which supply is growing at twice the long-run average, however, it is usually impossible for airline yield management analysts to raise prices to recover higher costs because the excess supply in the market drives the price down, not up.
Yield and revenue management systems of airlines generally try and predict, before the departure date, what demand will be in each fare class at the point of departure.
If the yield management system under-predicts demand, then the system will tell the yield management analysts to leave the cheap seats open for too long prior to departure, resulting in a depressed average yield. There is no doubt in my view that yield management systems have been under-predicting demand since the global financial crisis.
Variation in an airline's demand through time depends on two critical forces – the state of the economy and the capacity decisions of competitors. Neither force is incorporated effectively into the mathematical algorithms that drive yield management systems so it is not surprising that the demand forecasts of these systems go astray from time to time.
It is for this reason that airlines need good macro and micro economists to provide advice for the good of the airline (as opposed to non-airline economists who provide advice to airlines for the good of their employer). Only one of the domestic airline groups in Australia currently has an in-house economist.