Descriptively, there is certainly a taboo against racism and sexism. There seems to be moral emotions like anger, contempt and perhaps even disgust towards people expressing views percieved as prejudiced. However, I'm not sure any harm is necessary to explain why these terms are used. For instance, I believe most would have these emotions even in cases where there is no link to any real harm to any of these demographics. If I met a hermit in the woods expressing negative stereotypical views about certain demographics, it's hard to say what harm this hermit caused, yet I would still likely feel the same contempt as if I heard it uttered by someone in a social context. Thus, I don't think harm is the relevant factor why we have these responses.

If we are presented with a law, or an intersection of laws, that tell us how to differentiate between two acts with good consequences - but one is clearly superior due to the situation or context - isn't the intention just being displaced and projected onto a law or number of laws created by people that, if followed correctly, selects the best outcome based on a deficit of good consequences (its internalization dictates it must contribute to the maximum expected value, which implies that one outcome is preferable to another)? — ToothyMaw

If I understand you correctly, you are saying that when a rule is selected for internalization, the intention behind that choice would still follow (4) in your argument. That is:

4. Good intention requires a desire to bring about good outcomes — ToothyMaw

So, let's say there is a choice made between two rules, R1 and R2. R1 states: "One should keep one’s promises," and this would lead to 100 expected well-being. R2 states: "One should only keep promises made to those one likes," and let's say this rule only leads to 10 expected well-being. Then, at this moment of selection between R1 (100 good) and R2 (10 good), where they select R1 since it maximizes expected good, their intent would be like (4). And good outcomes are only the ones with the best consequences (from 8). Thus, a rule-consequentialist would be required to internalize R1 or they would lack good intention.

And if so, does the impartial defense of such rules not support the logic of my argument? — ToothyMaw

I have two comments on this. Firstly, I still think a rule-consequentialist following Hooker's theory would reject (4). Sure, the selection of the rules would be about maximizing consequences, but the intention behind following these rules would be something like "it is the most impartially defensible theory". Perhaps this is also due to how well it matches our convictions, it is consistent, gives explanatory power behind why certain convictions are right or wrong, and gives novel answers to moral dilemmas. The reason for being a rule-consequentialist would be that it is the moral theory that answers these kinds of questions the best. Thus, when selecting rules, the rule-consequentialist should select R1 rather than R2 because R1 leads to more well-being, but the intention would be "follow the most defensible moral theory" rather than "desire good outcomes." I think this is different enough to maintain my original point.

However, it is possible that by denying (4) the rule-consequentialists actually abandons consequentialism. I'm not sure about this, but I know Peter Singer has criticized Hooker's theory for being a form of deontology, only being consequentialist in name. If that is the case, then it would be no problem that Hooker-style rule-consequentialists reject (4) since your argument isn't even about them.

Secondly, even if the argument holds for the rule-consequentialist and (4) remains in the way that it is projected onto the choice of rules, I don't think it would lead to an absurd conclusion compared with act-utilitarianism for instance. Each decision wouldn't be determined by maximizing the good, rather, a limited number of rules would be followed. Everyone can only internalize so much, and the more complicated the rules, the more costly to internalize. The cost of internalization would also be affected by the cost of going against human psychology. If the rules require one to live a life of self-sacrifice, then the cost of internalization would likely be impossibly high. The resulting rules would by that token be far less demanding.

ANOTHER THOUGHT
I hope I'm not bombarding you with too much stuff, but I had another thought. How do you think your argument applies to ethical egoism? It could be considered a consequentialist theory as the "best outcome" would be something like: "maximizes what is good for oneself."

If you are a consequentialist, you must always have a desire to bring about the best possible consequences - not just good consequences - or you do not have good intentions — ToothyMaw

This isn't necessarily true if one subscribes to rule-consequentialism. For instance, Brad Hooker's defense of rule-consequentialism—that an act is wrong if it is forbidden by the code of rules whose internalization by the majority of everyone has maximum expected value—denies this psychology. Rather, he argues a rule-consequentialist could hold the following psychology: (1) their fundamental moral motivation is to do what is impartially defensible; (2) they believe that acting on impartially justified rules is impartially defensible; (3) they believe that rule-consequentialism is on balance the best account of impartially justified rules. Thus, I don't think all types of consequentialists must always have a desire to bring about the best possible consequences.

So, it seems that not only is consequentialism incredibly demanding, but it is also an exercise in self-deprecation if you are not some sort of selfless, Jesus-like figure incapable of a cheat day. — ToothyMaw

I believe this is one of the main reasons Hooker defends rule-consequentialism rather than act-consequentialism. He believes it avoids the demandingness of something like utilitarianism, which requires huge sacrifices from individuals.

Thoughts matter to some extent, but it is fuzzy to me. Some people have occasional violent thoughts about others, but this is clearly not as bad as actually doing violence. Maybe fantasies aren't bad in themselves, since they have a conscious unreality — the thinker is on some level aware that they are merely fantasizing. I think that at least mitigates most of the wrongness. Still, the nature of these thoughts can reveal something about one’s character, and different thoughts may carry different implications. Having vivid thoughts about doing something wrong could be said to be at least instrumentally wrong since it might dispose one to do the wrong thing, but would it be bad for its own sake? Take the following thought experiment:

The VR Prank. It's the year 2082 and Virtual Reality has developed drastically. As a joke, Adam plugs his friend Burt into a custom scenario without informing him. There is no visible change, but everything is now only a simulation in Burt’s unaware mind as the scenario plays out and Adam observes the events through the technology. Adam wanted to see how Burt would react if a life-threatening scenario occurred, maybe to rattle him a little bit. Thus, the representation of Adam collapses unconscious before Burt’s eye, laying helpless on the floor. Yet, instead of calling for help, Burt looks over his shoulder, picks up a pillow, and starts suffocating the unconscious representation of Adam. Horrified, Adam interrupts the simulation. Burt defends himself by saying it wasn’t real, it was only thoughts in his head and he didn’t cause any harm. Is Burt immoral?

I believe, yes. Since Burt thought it was real, I would say he did something as wrong as if he did it for real. Of course, Adam may also have done something wrong by doing this intrusive prank, but I think this doesn’t change the badness of Burt’s character. It would be reasonable to treat Burt the same way as if he had actually suffocated Adam in real life. At least, that is how I view moral responsibility: the consequences don’t matter. For instance, we can imagine that another person, Carl sleepwalked and unwittingly suffocated Adam. Here, the action would actually cause real harm, but I think Carl has done nothing wrong. Thus, it is not what one causes, but what one intends and thinks that determines how one should be judged.

To conclude, it seems to me that if there is a thought, but they think it is real, it should be deemed wrong in the same way. They are a person that is willing to do something wrong, and they should be treated as such. When it is a fantasy that one is aware of to be fantasy, I find it far less likely to be bad. Perhaps the degree of bad acts that is caused by such fantasies adds some level of wrongness. Still, I’m not certain if this is particularly notable. All of this may be merely instrumental. If Burt did evil things in a simulation that he would never leave, and he would never harm anyone, I’m not sure the thoughts would actually be intrinsically wrong. Yet, I think it is our intention that we should hold as the true markers of our ethical character, not just the physical actions we take.