Fails to Deliver are recorded at the DTCC and those fails are witnessed at the clearing and settlement end of a trade. The DTCC does not see how a trade is reported to the market long, short, short exempt, etc...but only sees whether a fail exists after a trade is executed.
thus, any DD that has taken place in which it concludes that short volume and how a trade is marked somehow masks fails to deliver would be dead wrong.
Interesting that you say this and yet the metric you use to justify your argument is the daily short sale volume. EVERY one of those trades represent tape transactions and cleared trades. If somebody were to try and mask their trades they would do so by hiding the trade from the public market.
Where you are in your DD is 2000 vintage. The original FTD data from back then showed massive FTD';s for long periods in time. This is 2012 and the data is reported to the public now and the results are significantly different. It is amazing that every scam has a full contingent of investors that want to blame NSS for the events taking place and NONE have ever proven to be correct.