The Chrysler debacle has given rise to another CDS-related claim: these dastardly products permit scenarios in which bondholders are willing to put a company into bankruptcy by distorting the investors' incentives. If the bondholders own a lot of CDS, then bankruptcy is more profitable than ongoing operations or restructuring.
There are two major problems with this argument. The first is that it is premised in the notion that bondholders, in the absence of CDS, would never choose bankruptcy over an alternative. This, history and common sense show, is just not true. Frequently, bankruptcy is a much better option that continuing operations (witness the airlines every few decades, as well as countless companies finding the current environment too stressful) because it provides current debt relief and perhaps more importantly may open financing channels that would be closed otherwise. Many people believe that Chrysler should have gone bankrupt years ago. The fact that auto companies' profits for two decades have been driven solely by their financing arms and not their manufacturing operations is telling. Who knows, a Chrysler that went through bankruptcy instead of a disastrous merger could be healthy today! So, the premise that "only with CDS" would bondholders choose bankruptcy over another option is absolutely ridiculous.
But even if someone accepts that CDS are not the sole reason bondholders choose bankruptcy, motive may still come into play. In the above examples, the bankruptcy choice was "good for the company" but not necessarily "good for the bondholder" (though it is very difficult for me to think of an example where that is actually the case, since by definition the bondholders own the company's assets). In a CDS-motivated bankruptcy push, one might argue that the choice is profitable for the bondholders but not necessarily good for the company. Since such a claim is not falsifiable - I can neither find a convincing case for or against it (though I lean strongly against in most cases) - I will instead point out that the situation is hardly unique to CDS. If I have a portfolio of two competing companies' bonds, and the bankruptcy of one will benefit the other, then as the weaker company approaches bankruptcy I may push it to file in order to reap the benefit on the rest of my portfolio. In other words, a long-only bond portfolio containing negatively correlated distressed assets is pairwise equivalent to a long and short credit portfolio. CDS does not introduce this scenario, it just makes it more obvious to the press that creditors hedge themselves.
It is a sad state when we blindly rebel against what we don't understand. If we were dealing with a CDO^2, I completely understand why there would be - and is - some nervousness about the nature of the product - because there simply isn't a straightforward way of defining the exposure. But CDS is not a complicated product despite its reputation. I frequently teach the basics of CDS and find, properly explained, that the product is much easier to explain than the dynamics of an equity investment (witness: leveraged ETFs). Companies blow up all the time on equity investments, derivative aided or not (Ackman's TGT fund, for example), and we don't hear the fallout, because we believe that stocks are "understandable" and CDS is "complicated," so the equity blowup is the investment firm's fault, but the credit blowup is the product's fault.
Remember, AIG didn't take itself down with incompetence, negligence, and greed... CDS did. And that's all there is to know.