First: citation needed in spammers using paid accounts. Why would they pay to spam, go through credit card fraud detection and so on when a free account is (by definition) zero cost? Mind you, I am talking about pure spam and not crazy sophisticated phishing scams. Those certainly exist in Twitter.

Second: scale. Twitter has millions of users, it is not profitable and is suing advertisers in a desperate attempt to justify its lack of revenue. Let's just say that a miracle happens and I get my dream number of paid accounts (10k at $29/year). The operation would be profitable enough to pay myself more than I ever made in any job and still contribute back to the downstream projects. I could literally close registrations on my service, or start a different vetting process.