Regulating Fannie and Freddie

[Adrian II]

This idea also informs my view of regulation going forward. Many people are calling for the government to regulate the procedures used by credit rating agencies. This is a horrible idea because the regulators have no logically sound procedure. If the government did oversee credit rating procedures, this would lead to an even more dangerous level of misplaced trust in flawed models.

Besides Banquo's Ghost proposition that we abolish incorporation, this is another highly original contribution to the thread. It is an approach (or rather, an issue) which never even occurred to me.

Question for you. Gaussian prediction is mostly used when we don't know the mechanism underlying a distribution, right? If we would know the mechanism, we might take better samples and reach more sound conclusions with regard to markets. This points to inadequacies in economic theory rather than statistical analysis.

Or am I being blond here?

[Oleander Ardens]

Mangudai brought up what brought down "Long Term Investment" roughly 15 years ago, and now so many big names. When you roll the dice each new roll has the same probability. Rolling 66 times a 6 is very very improbable.

But in our real social world a session of 4 sixes in a row can highly increase the chance of a six in the next roll, because people may expect this exit. In our highly interconnected world, especially in the financial one everything influences everything with a varying degree. Some seemingly unconnected events can shape the peception and create in the mind of the majority a very strong correlation between events.

This in turn can morph into a self-fullfilling prophecy like the famous bear or bank run. Many think Lehmann will fail, many sell, more see this as a sign of Lehmann failing more sell. While usually there are enough different opinions to moderate a stocks movement in highly volatile and nervous markets a rumour can bring down a titan. Something which is quite unthinkable in pure statistical analysis. But something which happens now.

P.S: LT Investment thought they could only loose 5 millions a day at most, and they had two Nobels in their ranks and other highly esteemed mathematicians making this calculations. Then they lost it at a rate of over 50 millions a day.

[Mangudai]

Besides Banquo's Ghost proposition that we abolish incorporation, this is another highly original contribution to the thread. It is an approach (or rather, an issue) which never even occurred to me.

Question for you. Gaussian prediction is mostly used when we don't know the mechanism underlying a distribution, right? If we would know the mechanism, we might take better samples and reach more sound conclusions with regard to markets. This points to inadequacies in economic theory rather than statistical analysis.

Or am I being blond here?

Here are two quick examples one is Gaussian the other isn't.

Line up 100 people and measure their height. The distribution is Gaussian.
Line up 100 authors and measure how many books they sold, one of the people is JK Rowling author of Harry Potter. Her score is thousands of times greater than everybody else in the line put together. Not Gaussian.

Automobile insurance is well modeled by Gaussian methods.
Home owners insurance breaks the Gaussian model when there is a hurricane or tsunami. (Companies like All State barely remained solvent after Katrina.)
AIG wrote insurance on financial derivates. At the start of the crisis the CEO said "Our balance sheet is bullet-proof". He was relying on Gaussian models.


Economic theory is inadequate and may always be so. Oleander pointed toward one reason why. Perception and what we think other people are going to do is central in markets. Right now the best bull market are stable earners like packaged food, toothpaste, toilet paper, etc. I'm up 8% on that part of my portfolio in the past few weeks. The sector will probably go up another 8% as everybody rushes in hoping not to be last. Then when it appears that everybody is in who wants in, the stock prices will suddenly drop back to their normal levels.

Sales of toothpaste are Gaussian. Sales of the toy most requested this Christmas are not.

[Crazed Rabbit]

A opinion article on how Fannie and Freddie became so top heavy they toppled:
http://www.bloomberg.com/apps/news?p...d=aSKSoiNbnQY0

Back in 2005, Fannie and Freddie were, after years of dominating Washington, on the ropes. They were enmeshed in accounting scandals that led to turnover at the top. At one telling moment in late 2004, captured in an article by my American Enterprise Institute colleague Peter Wallison, the Securities and Exchange Comiission's chief accountant told disgraced Fannie Mae chief Franklin Raines that Fannie's position on the relevant accounting issue was not even ``on the page'' of allowable interpretations.

Then legislative momentum emerged for an attempt to create a ``world-class regulator'' that would oversee the pair more like banks, imposing strict requirements on their ability to take excessive risks. Politicians who previously had associated themselves proudly with the two accounting miscreants were less eager to be associated with them. The time was ripe.
...
What happened next was extraordinary. For the first time in history, a serious Fannie and Freddie reform bill was passed by the Senate Banking Committee. The bill gave a regulator power to crack down, and would have required the companies to eliminate their investments in risky assets.

Different World

If that bill had become law, then the world today would be different. In 2005, 2006 and 2007, a blizzard of terrible mortgage paper fluttered out of the Fannie and Freddie clouds, burying many of our oldest and most venerable institutions. Without their checkbooks keeping the market liquid and buying up excess supply, the market would likely have not existed.

But the bill didn't become law, for a simple reason: Democrats opposed it on a party-line vote in the committee, signaling that this would be a partisan issue. Republicans, tied in knots by the tight Democratic opposition, couldn't even get the Senate to vote on the matter.

That such a reckless political stand could have been taken by the Democrats was obscene even then. Wallison wrote at the time: ``It is a classic case of socializing the risk while privatizing the profit. The Democrats and the few Republicans who oppose portfolio limitations could not possibly do so if their constituents understood what they were doing.''

Democrats love regulations, except when it's on the government. Pape made a good analogy about the football team.

CR