There is also a case to be made that even though initial investors took the risk to get a lower cap, they benefit from later investments even if they get a lower cap since the company becomes more valuable, making it more likely that they will actually profit more. While I agree that it's not ideal to have to allow later investors at lower caps, it could still be beneficial to earlier investors for the startup to accept these later investments. I don't think the moral burden should rest on the startup to go back to the earlier investors and offer a revised deal. Not only is it inefficient, but it also gives earlier investors the advantage when it comes to a hot startup that everyone wants to investment in initially. They can just bid with a high cap, guaranteeing that their offer is accepted, expecting the startup to come back because of it's "moral obligation" later on when other investors offer them lower caps.

In my scenario, where the second investor isn't a domain expert, but says something like "I will say yes to $xxx at a a lower cap", then does the startup have a moral obligation to give the same deal to the first investor? I think so, but could be convinced otherwise. Maybe this never happens.

It happens. There are investors who are notorious for offering lower caps to startups that have already started raising money. The solution is essentially to route around them. We advise startups to approach such investors last, when they've already raised enough that they feel comfortable saying "take it or leave it."

(There was a big kerfuffle a while ago when an email of this type got leaked.)

There's another case, though, when the startup has initially raised money at a higher cap than the market will bear. We warn founders about this, but they don't always listen. In that case they give the earlier investors the same lower cap that they negotiate with later investors.

This seems very inefficient for the startup. They have people willing to offer more money for less shares, so why not go down the list until they have enough money? Or are there second-order effects I am not seeing here.

It may not necessarily be inefficient. YC has an established brand. Initial YC investors know they will get the best possible terms, so they would be wise to offer the max value. The start-up it could take this "max" valuation to all future investors.

The alternative would be a low initial offer which would essentially keep anyone else from paying more. In essence the YC start-ups are giving the initial investor a money back guarantee on the difference between their offer and the lowest offer.

Are you suggesting that said moral obligation should be a part of the protocol?

I feel like it would make more sense for someone in that scenario to get advice for their specific situation, rather than spelling it out in the protocol.