alt Hacker News"refuse emails" already exists, that's the "hard bounce" thing that the OP is talking about.
Phishing and malicious links exists on every kind of messaging - email, forums, whatsapp, telegram, SMS... you name it, it has spam/phishing. You'd think that at least "real name" or "pay a bit per message" might discourage this, but example of SMS (text) shows otherwise.
With any system the server operator will be able to delete messages, and/or gatekeep information - unless you have some sort of "big brother" setup where every message goes through government-operated monitoring server, and I cannot believe anyone intentionally choosing such a system.
I can refuse a very specific email configuration, but there are so many simple variations, it isn't all that effective. Tons of cold email providers have you first buy 100 domain names, warm up the IP's with bots, and then send the same emails from a variety of email addresses/domains/servers. I think the question is does this low barrier to entry really outweigh the added challenge of allowing low volume actors host their own systems?
(steps on soapbox) And as long as we are talking about SPAM, why in God's name if I block a text message and then the phone number from ringing is there an additional block required to prevent voicemail? Oh, I see, it is so Verizon and others can charge me to block voicemails. Even on Google FI, if I block a number on Messages it doesn't carry over to calls which doesn't carry over to voicemail. Enshitification. (steps down)