Geo's right. The view that the lower fares equal lower revenues is overly simplistic.
To take it to extremes, if all coach seats were $1M each systemwide, even if every airline implemented it simultaneously, the only result would be that nobody would buy any airline tickets.
The demand is elastic with respect to price. WN has proven that it is very elastic. Once one understands this, one can better grasp the true complexity of determining the ideal profit point.
Quick business math lesson.
Revenues = Price x Quantity. However, Quantity is a function of demand, and demand is a function of Price. Therefore, Revenues = Price x Q(Price), where Q(Price) is the quantity demanded at a given price. To maximize Revenues, the first derivative must be zero and the second derivative must be negative. To find that price point, then, one must know the shape of the Q(Price) function, which is basically the elasticity of demand with respect to price.
But that's only how you would maximize revenue. The goal is to maximize profit, not revenue. Profit, of course, is Revenues - Cost. Cost is, in and of itself, a mighty complex function made up of static costs, unit variable costs, and other variable costs. So,
Profit = Revenue - Cost = [Price x Q(Price)] - C(Quantity)
Since Quantity, as we established above, is a function of Price, we can say that Cost is also a function of Price and ultimately can use the following:
Profit = [Price x Q(Price)] - C(Price)
Again, as above, to maximize Profit, the first derivative of the Profit function must be zero, and the second derivative negative.
That is how you determine the price point at which you maximize profits for your business. And that only works if the entire industry has the same cost function (one subset of which is a monopoly). Toss in the different cost functions for the different players, and each player will have a different profit-maximizing price point.
Ever wonder what yield management does all day? That's it.