Fed Deals in Lots of Junk at Primary Dealer Credit Facility

A preliminary analysis of the Primary Dealer Credit Facility (PDCF) source data released today by the Fed would seem to indicate a startling fact: Approximately 36 percent of the collateral posted, on average, was in the form of equity securities or junk grade debt.

Federal Reserve
Federal Reserve

Furthermore, only approximately 1.3 percent of the collateral, on average, was of the type traditionally posted at the Fed's Discount Window: U.S. Treasury or Agency Debt.

(Those numbers, however, are difficult to interpret—because collateral might be counted multiple times. John Carney explained earlier today: "Although the total numbers appear very large, the Fed never had anywhere near $8.95 trillion of loans outstanding under the program. The loans made under the PDCF were overnight loans, which were rapidly repaid or rolled over into new loans. This inflates the total number because the Fed counts each roll-over as a new loan.")

However, the general trend would seem to remain intact: The quality of the collateral posted for cash infusion from the PDCF appears to be rather weak.

If this discussion sounds familiar to you, it should. The PDCF collateral issue has been hashed through before.

On July 30, 2008 the Fed issued a press release stating that the PDCF program would be extended, "in light of continued fragile circumstances in financial markets."

The PDCF had been in existence since March of 2008, when it was created in the wake of the Bear Stearns collapse. The July press release from the Fed reiterated the rationale of the program: "The PDCF provides discount window loans to primary dealers"—essentially extending discount window operations to primary dealers of U.S. The discount window, of course, is a Fed mechanism used to provide liquidity to depository institutions, as opposed to investment banks.

Treasuries, which would not otherwise be eligible for discount window support.

It also restated—in the most unequivocal terms—the types of collateral that were eligible under the program. "PDCF provides discount window loans to primary dealers, collateralized by investment-grade securities."

However, on September 14, 2008, the Fed issued a press release detailing modifications to the PDCF program. The collateral eligibility requirements for PDCF were being changed. According to the release, the Federal Reserve had announced "several initiatives to provide additional support to financial markets, including enhancements to its existing liquidity facilities." (Emphasis mine.)

Here is the substance of the 'enhancement', according to the press release: "The collateral eligible to be pledged at the Primary Dealer Credit Facility (PDCF) has been broadened to closely match the types of collateral that can be pledged in the tri-party repo systems of the two major clearing banks. Previously, PDCF collateral had been limited to investment-grade debt securities."

No direct mention of junk-rated fixed income securities—or of equity securities—is made. Instead, the 'broadening'—or weakening—of collateral requirements is couched in the vague language of using securities "to closely match the types of collateral" that is eligible to be pledge "in the tri-party repo systems of the two major clearing banks".

This point wasn't lost on The Wall Street Journal, in an article published the following day. The Journal reported, on September 15th, that the Fed would be "taking a wider array of securities, including equities" after modifying the PDCF program.

We are now getting our first peek at what that new mix of collateral looked like—both before and after the changes were made to collateral requirements.

As I pointed out above, PDCF is a short term lending facility.

Therefore, it is highly likely that the same collateral would likely have been posted multiple times, effectively rolling over that collateral against multiple overnight loans. That almost certainly would result in the same collateral getting counted several times in the data released today.

What is the impact of counting the same collateral multiple times when attempting to analyze the mix of securities posted for PDCF? Using just the rollup numbers, it's impossible to determine the precise mix of assets at any given time. However, it seems likely that the totals I listed above would be at least somewhat reflective of the central tendencies of the assets posted as PDCF collateral.

It would then seem reasonable to wonder about this: Wouldn't the credit quality of the collateral deteriorate as the banks liquidity positions worsened? Following that line of thought through to its logical conclusion: At those times when banks were experiencing the greatest weaknesses in their liquidity positions, and likely posting the lowest quality collateral, wouldn't the balance sheet of the Fed be under the greatest strain?


Questions? Comments? Email us atNetNet@cnbc.com

Follow NetNet on Twitter @ twitter.com/CNBCnetnet

Facebook us @ www.facebook.com/NetNetCNBC