Money. It's not profitable to run an encrypted messaging service for free, because you can't extract any data from the chats, and people aren't willing to pay what it costs to maintain a good app.
It takes a few maintainers per platform and a few people willing to lead the effort, but I think we could get enough people on Hacker News to help out.
(I've actually been debating making an "Ask HN: Would you join this effort?" post but have been putting it off for one reason or another.)
https://blog.torproject.org/sunsetting-tor-messenger
"We, the people" is such a fragile concept when it comes to resource stability (time, money, dedication). Looking at a project like Linux, there are commercial interests that fund a large part of the pipeline.
Running the actual network and developing the various messenger clients add a layer of organizational complexity that, in my perspective, only private companies have been able to honor, and only to the extent described by the OP.
One could settle for less Service, but then we have XMPP with various encryption schemes. Too complex to set up.
The ideal would require no servers, because the chat would be fully peer to peer. People wanting to be anonymous could just connect through Tor. (Each user account is just a private key, etc, etc.)
Of course, in reality everything is behind a NAT and/or firewall. It wouldn't be too difficult to have "dumb" servers that just exist to pass messages hosted by the community though.
I otherwise can't imagine how to bootstrap a network.
Maybe users could pay to receive messages.
Pay to recieve a possibly unsolicited message?!
That's insane.
If a user's address is private tho, then asking them to pay to receive messages should be as sensible as the degree to which they protect the distribution of their address.
Facebook started with a narrow target: students of some uni. Look for another similar group of persons, that want privacy and don't mind spending a few dollars.
Edit: Whatsapp (everybody here in Spain has it) was a paid app at first. They gave one year away... eventually they made it free, around the time that Facebook bought it.
But you're still probably right overall. Marketing to businesses or teams within businesses is probably the best bet.
It turns out that the skills that the company has acquired by doing this pays back with consulting contracts.
(Sucks as defined by OP)