I think there's also evidence that the "short bubble" might be more widespread than we think. I'm not entirely convinced that Melvin was able to unwind their 50mln short position and drive the short interest down on Ortex (they're the best thing we've got in terms of guessing how much true short interest there is, because exchanges only report once or twice a month on a week or more delay) by covering. Equally likely in my mind is that Citadel bought Melvin's book thinking that it had a better plan than unwinding the short just yet. Instead, "hide" the short someplace that Ortex isn't looking and that the exchanges won't report on for another 3 weeks. That would be their order flow with a company they are in a unique position with - Robinhood - where lots of the buy orders are coming from. If you were Citadel and a little unscrupulous, you could immediately route every sell order on RH to the market but keep the buy orders for yourself. This positions you as short to Robinhood users, and you get to send those sold shares back to the book you bought from Melvin to unwind that short. So, the short does still exist, but now it's not visible to outside brokers. Outside brokers, when Ortex calls them up, will say, no no, Melvin covered a bunch today and short interest actually went down! So Ortex publishes a lower SI number. But, that short didn't actually just disappear. Either Citadel, or Robinhood, now are short to their customers.
Are there other companies that have "hidden" short interest in this way? Are some of the big unexplained market moves in companies without short interest some of these "hidden" shorts unwinding before everyone finds out?