The Need for Reliable Information

There has been much debate about whether the financial crisis is driven by lack of liquidity or from fears about lack of adequate capital and solvency, but I'm starting to think a third component is important as well, the complete breakdown of traditional information flows, and a loss of confidence in the models used to evaluate that information. Markets need information to work properly, and the information financial markets need is not available.

For example, investors can no longer trust what ratings agencies tell them. A crucial piece of information, information designed to break informational asymmetries between firms and investors, turned out to be unreliable. In addition, investors can no longer believe the numbers they see on bank books. The numbers might say the bank is solvent, but how reliable are those numbers?  And even if the numbers are meaningful today, will they be meaningful tomorrow? Is there any way to actually value the assets a lot of these banks have on their books when there is essentially no market for them, no way to engage in price discovery? Investors no longer trust analysts and the models they use. They watched the business channel dutifully and all they heard was about the gold mine in housing. Sure, there were a few voices on the other side, but they were in the minority and mostly marginalized. All that bullish advice about housing turned out to be wrong. And there's no reason investors should trust the models used to process information either. The models used for risk assessment turned out to be far wide of the mark - a costly deviation - and if you go back and look at the Fed's forecasts of coming economic conditions (or the forecasts coming from the regional banks), it's very clear the models were underestimating the severity and length of the downturn, enough so to be relatively useless. At a more individual, face to face level, I suspect their are many homeowners who believed what their real estate or mortgage broker told them are now wondering how they could have been so foolish. They won't believe them next time. They won't know what to believe.

As I think through each stage of the mortgage process and what has gone wrong, it seems to me that the traditional information flows that are needed for people to make economic decisions, especially risky ones, are no longer present, or if they are present, simply not believed. And without the information people need to make decisions, the markets freeze up.

It's the feeling you have when you suddenly discover that everything you thought you knew about something, something you believed and relied upon for years, is wrong (like when you find out something your parents told you just isn't so). Those are moments that can stop you in your tracks while you reevaluate and figure out what it all means, while you take time to figure out how you should respond in the future.

We have recognized that liquidity and solvency are problems, and we have directed policy to try to address those problems, but I am not sure we are devoting enough attention to repairing the collapsed information structure. You can get around the problem through government guarantees or other types of insurance, but those create other problems, so it's best to avoid this if possible. However, it's going to be difficult to convince people they can trust this information again, people won't easily believe a ratings agency, real estate agent, risk assessment model, etc. just because someone announces that the problems are all fixed now, models can't be repaired overnight, so on some fronts time may be the only real solution. But on other fronts, perhaps we can do better. This is not my area, so maybe what we can do is limited here too, but is there more that the government could do, for example, with accounting standards or required disclosures that would help people evaluate the stability of a particular institution? Are there changes that could be made to give buyers and sellers more confidence that the people acting as their agents in the transaction have the right incentives? Is there some way to immediately change the regulation and structure of the ratings agencies that can help to restore confidence in their assessments of risk? The point is that we need to move now to start repairing the problems that are limiting the availability of information needed for these markets to function.

Perhaps the most important thing the government could provide is confidence in bank balance sheets. There are lots of ways to do this, e.g. the government could purchase toxic assets through auctions, and the auctions would serve as value discovery mechanisms, the government could flood the banks with capital so that there was no doubt about their solvency, or it could simply put a price floor under some of the assets on the books, i.e. say that they stand ready to buy any and all of a particular class of asset at a pre-set price (heavily discounted). People could then put a lower bound on the value of the asset side of the balance sheet, and they wouldn't have to worry that the banks own actions or events outside the institutions control - an unanticipated failure of another bank that undermines a class of assets in its portfolio - won't suddenly change it's balance sheet position beyond a known amount. Somehow people need to be able to evaluate the bounds of the risks they are taking.

Big shocks don't necessarily shake the informational foundations of markets. There can be an event that occurs in the tail of the distribution of possible events that is viewed as just that, an unusual, costly event, but not one that fundamentally upsets our understanding of how the world works while at the same time undercutting the informational flows we use to understand these markets. I don't think the dot.com  crash, for example, caused us to question the reliability of the information we receive the way this episode has. After the crash, we still thought we understood how to use models to process reliable information. But this crisis has destroyed confidence in the information and the models we use, and it won't be easy to bring this back.

As noted above, while there may be some steps the government can take to help, solving this problem won't be easy, it will take time to repair the models and the information flows. That will eventually happen, but in the short-run the government must find some way around the problem. One way, the best way I can think of, is through insurance (e.g. the price floor above) and I hope we will see more movement along these lines. The deal with Citibank can be viewed as a step in this direction (there is a 29 billion dollar deductible and a 10% copay in the insurance they were provided - see the update at the end of this post), but more can be done - more must be done - to overcome the lack of reliable information in these markets.