SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Strategies & Market Trends : The Epic American Credit and Bond Bubble Laboratory

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: russwinter who wrote (23571)12/18/2004 11:27:52 PM
From: russwinter  Read Replies (1) of 110194
 
Jesse at Prudent Bear's comments on my duration shift theory:

Very interesting theory and threads jessel
NEW 12/18/2004 4:51:50 PM

Its possible, but some items don't lend credence to it.

I looked back over the specific structure of the Fed's holdings of US Treasuries, and there was little to no growth on the longer end of the curve throughout this period.

I suspect that buying has been going on the long end of curve by central banks, depressing the yield curve, but primarily in the custodial accounts which do not provide breakdowns. This would tend to support that the Central Banks are over-committed on the long durations.

However, it does not look like the Fed has been taking them off their hands.

I am looking at Section 3. Maturity Distribution of Loans and Securities in the Fed's Weekly H4.1 report.

I also did notice that for the Securities Lending, the majority of it seems to be ten years out or so. This is what causes me to suspect that the primary dealers are selling the long end short to the CB's in the custodials, and then covering as the yield flucuates, borrowing inventory from the Fed in the interim. Even though Security Lending is for a 'day' in fact if you track it they are 'rolling' those loans for multidays, taking them down as inventory is covered.

But, until we get inventories of the Custodials, there is no way to 'prove' this.

Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext