but it may also mean far less chance to innovate production practice.

If vertical integration is always good, the ideal situation is a single producer for any kind of product and no suppliers, which doesn't seem to work too well either.

Efficiency and security are always traded for one another. Unfortunately, markets will tend to the short term benefits so the pendulum will swing towards efficiency until it’s gone too far and something causes it to swing back.

Up to a point. But when you get to the point that an industry is so vertically integrated that only one entity can viably compete, you have a monopoly.

There's no monopoly here, there's a duopoly. Airbus and Boeing are in fierce competition with each other. On top of that, the Chinese are entering the fray too.

Efficient market theory would suggest that there is no margin in equilibrium, and what looks like margin is actually the cost of capital (which all companies, big and small, have to pay).

That's incorrect - you would only expect no "margin" in a perfectly competitive market, which obviously isn't the case in the market for airliners.

In markets where firms have market power, there is an explicit and well-studied phenomenon known as "double marginalization"[1][2], which is what the parent comment is describing. Vertical integration is one possible solution to this problem, although there are other approaches as well.

Double marginalization also occurs elsewhere in the air travel industry. Imagine I flew on American Airlines from Charlotte to London, and then continued on British Airways from London to Lyon. If both airlines set their prices independently, then they would both add a markup, and the combined price of the itinerary would be higher than the profit-maximizing price that a single airline would set. In this situation, antitrust authorities will often allow the airlines to form a joint venture and coordinate on price-setting. This reduces horizontal competition (because e.g. AA and BA are no longer competing with one another for passengers on JFK-LHR), but can provide consumer benefits if eliminating double marginalization allows airlines to set their prices lower on multi-airline itineraries.

[1] https://en.wikipedia.org/wiki/Double_marginalization

[2] https://mru.org/courses/development-economics/double-margina...

Economics is laughably bad at predicting real-world economic behavior.

That's too broad a statement. There are some theories in economics that work very well, like supply and demand, or arbitrage theory. The idea of the cost of capital is also a fairly well-trodden one, it is taught in business courses, used on wallstreet, and seems to work in practice.

I can't think of a single broad economic axiom or theory that doesn't depend—wholly or in-part—on the principle of ceteris paribus, which excludes their practical application to essentially everything in real life.

In a few cases, economic theories do a passable job of roughly mirroring real-world behavior, but that's the exception and not the rule.

>on the principle of ceteris paribus, which excludes their practical application to essentially everything in real life.

When a principle is violated marginally, conclusions made based on the principle are themselves violated marginally. You can't write off chemistry because relativistic effects violate the conservation of mass any more than you can write off economics because no two soup cans are identical.

Economics waves off secondary effects at an entirely different level than Chemistry.

Except efficient market theory is a bunch of bullshit that does not fit reality.