By way of analogy, if I create a new kind of medical device, the FDA does not say, "We don't have any regulations that cover that, so it's completely unregulated until we see the need to write some." No way.
But a new financial product, while it can't kill anyone, it can still cause massive damage. (We just got a case study in this in 2008.) The default for any new financial idea should be regulated, not unregulated.
And "but it's not a bank!" doesn't cut it. Mortgage securitization wasn't a bank activity, but it still nearly destroyed the world economy. Repo isn't a bank activity, either, but it played a significant role in the crash. Both are "bank-like" enough that they needed serious regulation. Since the risks were not yet understood, any regulation in place was not nearly stringent enough.
A great writeup about this by someone a bit more credible than Krugman can be found at http://blogs.wsj.com/economics/2010/02/23/so-what-exactly-ca...
Written by Yale and Wharton Professor Gary Gorton, who has held positions at the Bank Of England, the Federal Reserve and the FDIC.
"Innovation" is almost always another way of saying "We found another way to gamble with the customer's money."
The regulators, rating agencies, mortgage lenders, government, GSEs all contributed to people overleveraging, but thats not even the real issue.
The real problem is that the economy has changed, and all but the high skilled jobs are going overseas or being done by machines.
Bankers are not going to jail because they didn't break any laws. People need to stop being parrots and making claims about things they dont understand.
You look at the software industry, what do we see? Mild amateurs not encrypting personal data, or not protecting their sensitive data against hacking 101 attacks (we still see sql injections attacks in this day and age!). You see a very large shady industry prospering on invading the privacy of unsuspecting users (and not just the adclick of this world: google, facebook!). In a corporate environment, developers have a similar bad reputation than house builders in term of creating projects that very often take a lot longer that expected, fail, or simply are not fit for purpose.
Does it have no consequences? It has massive consequences and the worst data leaks are probably yet to come. Should we impose heavy regulations on the software industry and curb toxic innovation? It is certainly not an absurd debate. Obviously most people in the software industry will react to this with hostility. They will argue it would negatively affect innovation and progress, that there are good and bad apples, and that one cannot judge an entire industry without a minimum understanding of how it works and how it benefits the economy.
Well guess what: it is the same with the financial industry!