Assessment by consequence is a valid approach to evaluating the credibility of the applied ethics of an action; the problem is that consequence is by definition a post hoc evaluation of the rightness of an action. You can attempt a prediction of the consequences of an action, but as the o.p. notes, any action sufficiently complex to involve more than one other party, or has ramifications beyond the immediate consequences, is too vast in potential bifurcations to make a perfect prediction.
Assessment by consequence also leads to the other sticky ethical problems of ethical relativism and the comparison of unlike consequences. That is, you may view the consequences of an act as being harmless (i.e. stealing a pack of gum from a busy newsstand) but the owner may have a different view of harms (especially if many people do this.) This disconnect in common basic ethical stances gives rise to (ethically) irresolvable conflicts between parties which means that a third party must intervene to enforce a (presumably arbitrary) standard of behavior which may or may not be in concordance with one or both parties. This is essentially an abrogation of the concept of a “social contract” which underlies a free society.
Comparison of unlike consequences is even worse; if an action results in bad consequences either way, who is to say which is better? The burglar would argue that his act of theft does not justify the use of lethal force to inhibit or interfere; the home or business owner might feel, on the other hand, that the consequence of losing valuable property purchased with the labor of unrecoverable hours of one’s life or the apparent threat posed by an intruder fully justifies the consequences of the threat and application of lethal force. Who decides what consequence is most appropriate, and on what non-arbitrary, non-biased basis?
Assessment by consequences does not offer a unique and invariant system for assessing the potential ethics of a situation; it is a recursive syllogism with the minor condition stating that the consequence is ethical if it is not harmful (by the reasoner’s evaluation), and then concluding that the action was ethical. Of course, any real world application of a system of invariant ethics will always have extreme conditionals or exceptions; a doctor may elect to perform an operation which may possibly kill the patient in order to treat a life-reducing ailment. But “consequentialism” lacks any firm basis for evaluation.
In the context of the original argument (specifically, the purchase of one movie ticket to enter the theater to see another) it is asserted by some that the consequences make the action ethically permissible by way of satisfying the agreement between the cinema and the purchaser to pay a certain amount in order to enter the cinema, and that the cinema sees no consequences from the purchaser’s subsequence action to view another film. Aside from the fact that the contract is, in fact, not simply to enter the cinema and see any film but rather to see that specific film at a specific showing time (making this clearly a contractual violation) the purchaser is also unaware of the specifics of the agreement between the distributors and the consequences should the distributor discover that they are not receiving due revenues for viewings of the film, and therefore incapable of addressing the consequences in a holistic manner. (This is not purely a theoretical exercise; distributors sometimes send representatives to count the number of people entering a viewing and compare this to stated revenues; a cinema that has too much variance may experience penalties as a result.)
Of course, the reality is that cinemas don’t make their vig from ticket revenues (which at best offset the cost of showing the film) but from concessions, and so as long as you buy your popcorn and soda they don’t really care much which film you see. Distributors factor in a certain (small) proportion of losses, and don’t really care (much) if a few kids watch Star Wars several times in a row. In general, the harmful consequences from this are typically low, which is why little or no effort is made to prevent it except in limited circumstances; the cost of trying to do so is less than the savings from preventing this behavior. This cost vs. benefit analysis is hardly the basis for an ethical standard; it is pure fiscal pragmatism, an acceptance that some people will not be ethical, but that the costs can be absorbed.
Someone wishing to make an extreme ethical position can invoke hedonistic or even solipsistic consequentialism, i.e. I don’t care what the consequences are for someone else, or it doesn’t matter what the consequences are for anyone external to me. That is a perfectly valid ethical argument to make, provided that you are comfortable with being a sociopathic asshole straight out of a Neil LaBute play. Such a view does not coordinate with societal ethics, though; that is, if your personal philosophy is that other people do not matter, then you have not basis for indignant when other parties present you with the same argument.
Morality, as contrasted with the philosophy of ethics, is a somewhat different and more nebulous concept in which often arbitrary (or seemingly arbitrary) normatives are imposed by an outside authority, with no direct basis in consequence external to that authority. An act might be perfectly ethical under any rational circumstance (i.e. protected sexual intercourse between two consenting adults) but in violation of some applied moral code for reasons beyond consequence or rationality.
Stranger