Bob Ivry of Bloomberg sifted through documents obtained through a FOIA request, and his report, released last week, offers some thorough detective work. Yves Smith and Felix Salmon picked up the thread, and Dave Altig of the Atlanta Fed responded.
Discussion thus far has missed the big picture, and the issues raised are a distraction from the real problems that are revealed by this data.
First, about that 0.01 percent loan. The figure comes from the Fed’s data on its open market operations. (I have retrieved the relevant transactions and put them here.) The stop-out rate, the lowest rate at which loans were made, indeed reached 0.01 percent, or 1 basis point, in the final $20B operation. At the time, discount window lending would have cost you 50 basis points. So the Fed has, one might argue, foregone 49 basis points of interest on a loan. This loan must have been for less than $20B, since the average rate was higher. But let’s use $20B to get an upper bound.
The interest on a 28-day loan of $20B at 49 basis points is $7,622,222 (that is, 0.0049 * (28 / 360) * $20B).
The real story here is much bigger than eight million dollars.
What were these loans all about? The Fed was acting as lender of last resort, not to banks, but to securites dealers. This was a major shift in the Fed’s thinking, a shift that it made only reluctantly, and only when it felt it had no other choice.
Between March and September 2008, the Fed supported the dealers with the special liquidity programs PDCF and TSLF. PDCF provided overnight loans (of money) to primary dealers, while TSLF made one-month loans of securities.
Like TSLF, single-tranche open market operations provided term loans; like PDCF, ST OMO provided loans of money (and not of securities). The three should be grouped together, as the Fed did itself when it released its discussion of 2008 open market operations in January 2009.
The Fed had to support securities dealers, because in the decades leading up to the crisis, the financial system had changed. Tradable securities (MBS, for example) came to play a role that bank credit had once played. Dealers are indispensable in such a system, but the Fed did not have a ready-made way to support this function.
Altig draws a line (starting at “Another aspect” in his post) between ST OMO on the one hand, which he considers to be business as usual, and TSLF and PDCF on the other, which “relied on the authority granted under ‘unusual and exigent circumstances’ by section 13(3) of the Federal Reserve Act”.
But this claim ignores the rationale underlying all three programs, which was genuinely new regardless of the formalities. As the Fed itself reported, single-tranche open market operations were ” intended to…provide the primary dealers a steady financing source for Agency MBS” (see p. 11). But normal open market operations are about affecting the reserve balances of banks, with dealers making markets in which the Fed can intervene. ST OMO, like PDCF and TSLF, was specifically and explicitly about supporting dealers, which was not at all business as usual.
To be clear, I am not defending the Fed; it is completely right to find fault with the Fed’s handling of the crisis. But those who want a safer financial system should choose their lines of attack with some care. The Fed’s failure here is not eight million dollars of subsidy. The failure is that, prior to the crisis, the financial system changed, but the Fed didn’t. The failure is that, three years on, it still hasn’t.