Did A Lack Of Data Cause The Crisis?

I generally describe the financial crisis as a sort of perfect storm of things gone wrong. The housing bubble was certainly the catalyst, but then other factors took over. Banks had unknown leverage, securities were owned by Wall Street and investors that weren't well understood, etc. Karen Dynan, Vice President and Co-Director, Economic Studies at the Brookings Institution recently made a presentation (.pdf) that suggests one of the major problems was data. She argues that if more, better data had been available, then we might have known a housing bubble was being inflated and stopped the problem more quickly. I disagree with her particular argument, but agree with the general idea that more data is good.

Let me provide an example of the kind of thing Dynan is calling for. Her first point reads: "Developing and publishing aggregate indicators of financial imbalances would be helpful but will take some work." Here, she provides an example of a graph showing the ratio of home prices to rents:

Dynan 2009-12.PNG

The argument goes something like this: "Look! If we'd seen graphs like this then we'd have known that something was terribly wrong in the housing market. We could have then prevented the bubble from growing so large!" That sounds like a great theory, but having lived through the bubble, I think it's a touch naïve.

The reality is that plenty of people knew that a bubble was forming. Heck, way back in 2005 I even bought a shirt joking about the growing housing bubble (I have a thing for funny, nerdy t-shirts). And I was hardly a master-of-the-financial-universe. It was easy to see data on how much home prices had increased. In 2006, I remember creating graphs about how incredibly large the mortgage- and asset-backed securities market had grown -- even outpacing corporate debt for the first time.

I don't think data was the problem. Anyone who suspected that a bubble might be forming could easily find some data to back up the claim. The problem is that no one really much cared. Mortgage companies claimed ignorant optimism: "Sure these products seem wacky and the returns unsustainable, but it's real estate! It will be fine!" Homeowners sung a similar tune, never imaging that their home could lose value: "If we can't pay our mortgage after reset, we'll just sell the house for a profit, since it's bound to appreciate!" Wall Street also wasn't worried: "Oh c'mon. Those mortgage-backed securities are AAA-rated. Besides we have CDS underwritten by that giant insurance company AIG to hedge our exposure."

The signs were all there. They were just stubbornly ignored.

Dynan sort of anticipates this criticism. She characterizes this as "group think" to purposely fail to see the reality of a bubble:

One might argue that the most important need is for the relevant policymaking agencies to have such data. While this may be true, there are also arguments for making the information available more broadly. For example, some commentators have argued that regulatory agencies faced analytical limitations and tendencies towards "group think" that contributed importantly to their failure to recognize just how much risk was building up during the credit boom. To the degree that such arguments are valid, there is a case for providing greater information to the public so that there can be a more vigorous examination of and debate over financial trends.

So if the public had seen this data, then they would have objected to the housing boom. I think Dynan may have been at a think tank too long if she really believes that the public would understand -- much the less take seriously -- such information. Does she really believe that average Americans who vote more for "American Idol" than national elections will seriously heed economic data? And how would this work? Banks certainly aren't going to warn consumers -- they're too busy making billions of dollars off the bad mortgages they're writing. Would the government mail out pamphlets warning Americans of a housing bubble? Few pieces of junk mail will have found their way so quickly to the trash.

Perhaps I'm just hopelessly cynical, but I don't believe the problem during the crisis was that we just didn't have enough data available or well-understood. It was that nobody cared, because they were too enamored with feeling wealthier.

With all of this said, I should take a step back. More data is a good idea. I wish we'd known more about the magnitude of hidden leverage at banks. I wish we'd known the kind of potential exposure that AIG was developing with the CDS it created. I wish investors would have actually taken some time to understand the data behind the mortgage-backed securities they purchased, rather than just glancing quickly at the AAA rating as they forked over their cash.

But none of those problems really had anything to do with the housing bubble. The problem with bubbles is precisely that they're based on irrational exuberance. So the idea that investors or the public can rationally talk themselves out of a bubble is nonsense due to bubbles' very nature. What can help, however, is having more detail about the financial industry. That way, even if a bubble does form, at least the financial industry can be more knowledgeable about its state so that it can sustain the associated bubble losses without causing an economic catastrophe. So the focus shouldn't be on more information to avoid bubbles, but on more information to avoid systemic risk. That way, bubbles won't be so destructive.