John B. Taylor and John C. Williams write,
the spread on August 9 was 25 basis points above the pre-August 9, 2007 average. That is 7 times the standard deviation before August 9—more than a 6-sigma event. The mean through March 20 was 16 standard deviations above the old mean, which under normality would have been an extraordinarily improbable event.
The spread to which they are referring is the London Inter-bank Borrowing Rate (LIBOR) over what they call the Overnight Indexed Swap rate–think of it as the Federal Funds rate. This large spread, and similar spreads of historically low-risk assets relative to Treasuries, are the essence of the credit crunch that the Fed is trying to solve with novel balance sheet moves, such as the Term Auction Facility (TAF). On the latter, the authors write,
The model has two implications. First is that counterparty risk is a key factor in explaining the spread between the Libor rate and the OIS rate, and second is that the TAF should not have an effect on the spread. Since the TAF does not affect total liquidity, expectations of future overnight rates, or counterparty risk, the model implies that it will not affect the spread. Our simple econometric tests support both of those implications of our model.
The Fed is very big in the eyes of journalists. But in the context of world markets, the Fed may be too small to affect credit spreads.
READER COMMENTS
Gary Rogers
Apr 24 2008 at 1:04pm
When results go beyond the six sigma limit it means something significant has changed in the system. Does anyone know what this is? We know that the symptom can be described as “some consumers are over their heads and cannot afford to pay their mortgages” but the government response is the same as it is for any economic downturn, “consumers need to spend more.” If the change is not explained along with appropriate corrective measures, we will continue to see inappropriate responses from legislators.
Matt
Apr 24 2008 at 10:47pm
Lotta reading, but I have been following the LIBOR spread issue. I think the best summary is that the peak of the housing cycle was reached, we all expected that, but the rate of rebalancing the portfolios of the bank went faster than anyone expected. The Fed did its job as chief accountant.
Why the more rapid change in housing?
My best guess is that the loan aggregators hung around too long, they should have jumped ship two or three years ago. Its this hedging system, the hedgers never know when to quit and so they bring on higher rate changes than the system expected.
Why is this? A better question, is can we identify, at least, estimate how often an industry needs a mid-level financing institution, and how often it doesn’t?
We are going to always have this problem which I will state. The larger an industry is, the greater probability it will operate as a rank N+1 organization, vs a rank N. That is, it sometimes needs a secondary market, and sometimes not. The two possibilities are exclusive.
If my industry needs a second level redistribution, today, then tomorrow when I don’t need it, then, if the extra level stays anyway, I am stuck with an incomplete marketing level in my industry.
We can define an optimum hedging system for any market, but it is not regulation; which just installs the secondary market permanently.
We need a system that detects when activity in the housing finance business reaches a heightened level of activity, then the repackagers jump in, operate for a year or so sanitizing the debt. Then we want the secondary repackagers to dissappear, completely, as if they were never there.
The system needs to be prepared for state changes, we have to be able to add or remove secondary markets much faster.
Matt
Apr 24 2008 at 11:07pm
Thermodynamics and economics
and Paul Samuelson and William Gibbs?
I hadn’t a clue, and someone should have mentioned. Nothing I say is new, and all of it was accomplished 40 years ago by Paul working with William. Paul got the Nobel Prize, but until today I never looked at why.
I am not an economist, and Gibbs, to me, was just some scientist somewhere. I had no idea.
Paul, if he sticks around a few more years, should be able to complete the Theory of Quantum Mechanical Economics. His paper showing the efficient market theory leads the way by proving that cyclic prices still cannot be predicted from moment to moment. Paul knows where this theory leads.
Comments are closed.