Matt Taibbi is apparently back on the finance beat with a story on the collapse of Bear Stearns and Lehman that goes "deep into the weeds" of naked shorting. It's not clear to me why, since the only academic paper I'm aware of that actually studied the question found that the volume of shorts and naked shorts intensified after the bad news that caused the stock to plummet, not before, and at any rate, was never done in a large enough volume to cause price declines of the magnitude that we saw.
Taibbi seems to be more worried about the moral offensiveness of the practice than its actual impact:
The real significance of the naked short-selling issue isn't so much the actual volume of the behavior, i.e. the concrete effect it has on the market and on individual companies -- and that has been significant, don't get me wrong -- but the fact that the practice is absurdly widespread and takes place right under the noses of the regulators, and really nothing is ever done about it.
It's the conspicuousness of the crime that is the issue here, and the degree to which the SEC and the other financial regulators have proven themselves completely incapable of addressing the issue seriously, constantly giving in to the demands of the major banks to pare back (or shelf altogether) planned regulatory actions. There probably isn't a better example of "regulatory capture," i.e. the phenomenon of regulators being captives of the industry they ostensibly regulate, than this issue.
There are much better examples of regulatory capture, even in the financial industry, such as the SEC's ban on shorting financial stocks last summer . . . but that is neither here nor there. Naked shorting has benefits as well as drawbacks--it enhances the efficiency of the market in price discovery, as well as its liquidity. It is often done by exchange market makers who fill orders on the fly, and then hunt down the actual stock later, in order to keep orders flowing. It is also, of course, hated by CEOs, who like to blame evil short sellers, rather than their own mismanagement, for driving down the stock price.
In theory, a "bear raid"--selling a flurry of shares in order to push the stock price into a downward spiral--is possible. In practice, counterattacks are possible, making this a very risky strategy for shorts who can end up bankrupt if they can't find shares to buy at a reasonable price. It's also difficult, with a large and liquid stock, to sell enough volume to permanently depress the price; your counterparties start wondering where you're getting all this stock to sell. That's why, while there's pretty decent evidence that shorts, and naked shorts, can speed the mean-reversion of overpriced stocks, there's a lot less evidence--virtually none--that it can cause stocks to become underpriced for any length of time.
Even if there were evidence for successful bear raids, Lehman's creditors had ample reason to worry without a decline in the stock price. By the time the volume spiked, Lehman's fate was already sealed; either they were going to find a buyer, they were going to get a bailout, or they were going to bankruptcy court.
So why should this be priority #1, or even #30, for the SEC? Obviously, CEOs do not like any practice that speeds up negative price discovery in their stock, but this is not supposed to be the SEC's concern. There's legitimate reason to punish people for failing to deliver--after all, they're in breach of contract. But this is a problem for the contractees, or the exchange, not the SEC.
Update: Someone in the comments asks if I'm not conflating regular shorting and naked shorting. Answer: no; the paper I linked deals specifically with naked shorts, and finds that at least since the introduction of Regulation SHO in 2005, they have functioned primarily as a liquidity enhancer and a price discovery mechanism, rather than a market manipulation mechanism.
Our results are in sharp contrast with the extremely negative pre-conceptions that appear
to exist among media commentators and market regulators in relation to naked short-selling. While unregulated naked short-selling could be potentially manipulative, and the associated settlement failures could be somewhat disruptive to the smooth functioning of financial markets, the duly regulated naked short-selling that has existed after Regulation SHO appears to havebeen net beneficial for pricing efficiency and market liquidity, and Regulation SHO also appears to have successfully curbed the impact of manipulative naked shorting, and this reduction in the impact of manipulative naked shorting has continued through the 2008 financial crisis.