PrefBlog

Bloomberg has reported:

For the first time, the market will have a committee of banks and investors making binding decisions that determine when buyers of the insurance-like derivatives can demand payment and could influence how much they get, industry leaders said yesterday at a conference in New York. Traders also will revamp the way the contracts are traded, including requiring upfront payments to make them more like the bonds they’re linked to.
…
The plan doesn’t change contracts traded in Europe.
…
In one of the most noticeable changes for traders, those who buy protection will pay an upfront fee depending on current market prices, and then a fixed $100,000 or $500,000 annual payment for every $10 million of protection purchased. Now, upfront payments are only required for riskier companies, and the annual payment, or coupon, on most contracts is determined by the daily market level.

Dealbreaker, bless its heart, is contemptuous:

We are looking forward to the world where the only finance products permitted go up forever, and where everyone makes above average returns.

I’m mainly confused, and hoping that the Bloomberg reporters simply got it wrong. A standard up-front fee for the buyers of protection? It makes no sense. The buyer’s risk is limited to the sum of the present value of the payments required. A five year contract with a 5% premium limits the buyer’s potential loss to 25% of the notional value. The tail risk of the contract is owned by the seller of the protection, who can lose 100% of notional on the very first day the contract is in existence.

Selling protection has capital implications roughly equivalent to owning the bond. I have no problem with the idea that CDS sellers post margin equivalent to what they would have to were the position an actual bond – but most of them do already. The trouble started when the AIGs and MBIAs of this world sold protection without posting collateral or taking a high enough capital charge for regulatory purposes.

And what’s this with a fixed standard rate of $100,000 or $500,000 per year for the premium (paid by the buyer to the seller) on $10-million notional? That’s 1% and 5%, respectively. I can see that it might be very useful to standardize tick sizes, so that all contracts will trade with, say 10bp ticks … but those premia don’t make any sense.

Note that with 10bp ticks (any size ticks, actually), virtually every contract would have a value at the opening, which would be settled by cash payment between buyer and seller at the time the contract is written. I don’t have any problems with that idea – it would make contracts more fungible, particularly if settled by a clearinghouse.

But we are in the hysteria phase of CDS demonization and the politicians need a pulpit. ISDA has released a mild demur – I can only hope they’re more vociferous behind closed doors.

It was announced yesterday that JPMorgan’s analytics will go open-source; but no details of licensing have yet been released.

Assiduous Readers will remember the puzzle of the Effective Fed Funds Rate. Fed Funds were trading at 0.25% at a time when excess balances were earning 1.00% from the Fed, which appears to allow a risk-free arbitrage.

A recent Fed Press Release and its attachment may provide a piece of the answer. First, a little terminology gleaned from the attachment … a small bank does not need to satisfy its reserve requirements directly with the Fed. It can deposit the necessary funds in a (presumably bigger) bank’s Fed account, in which case the small bank is the “respondent” and the big bank with the Fed account is the “pass through correspondent”. Got it? OK:

As noted above, Regulation D currently deems the entire balance in a pass-through correspondent’s account at a Reserve Bank to be the exclusive property of the pass-through correspondent and to represent a liability of that Reserve Bank to the pass-through correspondent exclusively. Therefore, the pass-through correspondent must show the entire balance in its Reserve Bank account on its own balance sheet as an asset, even if the balance consists, in whole or in part, of amounts that are passed through on behalf of a respondent.

Accordingly, when a correspondent’s respondents want to earn interest on excess balances by leaving them with their correspondent (which in turn passes those balances through to the Reserve Bank), the correspondent has a larger balance at the Reserve Bank. As a result, the correspondent has more assets on its balance sheet and a lower leverage ratio for capital adequacy purposes.

In contrast, when the correspondent sells the respondent’s federal funds on the respondent’s behalf, the respondent directs its correspondent to transfer funds to the entity purchasing federal funds. This transaction is effected by a debit to the correspondent’s account at a Reserve Bank and a credit to the purchaser’s account at a Reserve Bank. On the correspondent’s balance sheet, all other things being equal, the correspondent’s assets decline (as does its liability to its respondent) because the correspondent’s account balance at the Reserve Bank is lower and therefore its regulatory leverage ratio would be higher.

Since the implementation of interest on excess balances through the October interim final rule, the actual federal funds rate has generally averaged significantly below the interest rate paid by the Reserve Banks on excess balances, although this spread narrowed significantly after the FOMC established a range for the federal funds rate of 0 to ¼ percent on December 16. When the market rate of interest on federal funds is below the rate paid by the Reserve Banks on excess balances, respondents have an incentive to shift the investment of their surplus funds away from sales of federal funds (through their correspondents acting as agents), and toward holding funds directly as excess balances with the Reserve Banks, potentially disrupting established correspondent-respondent relationships. A correspondent could offer to purchase federal funds directly from its respondents and hold those funds as excess balances at a Reserve Bank; however, such transactions could result in a significant reduction in regulatory leverage ratios for some correspondents. The Board believes that the disparity between the actual federal funds rate and the rate paid by Reserve Banks on excess balances may partly be caused by the leverage incentives imposed on correspondent institutions to sell excess balances into the federal funds market rather than maintaining those balances in an account at a Reserve Bank.

To address this problem, the Fed is proposing to establish a new account type, an “Excess Balance Account”.

These excess balance accounts (“EBAs”) would contain only the excess balances of the eligible institutions participating in such accounts, although the participating eligible institutions (“EBA Participants”) would authorize another institution (“EBA Agent”) to manage the EBA on their behalf. The authorization of EBAs is intended to allow eligible institutions to earn interest on their excess balances at the excess balance rate in an account relationship directly with the Federal Reserve Bank as counterparty without disrupting established business relationships with their correspondents. Continuing strains in financial markets and the configuration of interest rates support the implementation of EBAs; however, the Board will evaluate the continuing need for EBAs when more normal market functioning is restored.

EBA Balances will not gross up the pass-through correspondent’s balance sheet.

Update, 2009-2-2: Economic Policy Journal commented in early December on a December 1 speech by Bernanke in which he noted:

In practice, however, several factors have served to depress the market rate below the target. One such factor is the presence in the market of large suppliers of funds, notably the government-sponsored enterprises (GSEs) Fannie Mae and Freddie Mac, which are not eligible to receive interest on reserves and are thus willing to lend overnight federal funds at rates below the target.
…
Banks have an incentive to borrow from the GSEs and then redeposit the funds at the Federal Reserve; as a result, banks earn a sure profit equal to the difference between the rate they pay the GSEs and the rate they receive on excess reserves. However, thus far, this type of arbitrage has not been occurring on a sufficient scale, perhaps because banks have not yet fully adjusted their reserve-management practices to take advantage of this opportunity.

I am happy to see my ‘bureaucracy explanation’ front and centre!

ISDA has announced:

that J.P. Morgan has transferred to ISDA its CDS Analytical Engine. The CDS analytical engine, originally developed by the Quantitative Research group at J.P. Morgan, is widely used in the industry to price CDS contracts. ISDA will make the analytical engine available as open source code, thereby increasing transparency around CDS pricing.

“J.P. Morgan has invested a lot of intellectual capital in this analytical engine. Its willingness to assign this to ISDA for us to make it available as open source to the entire industry demonstrates our collective commitment to the integrity of the CDS product,” said Robert Pickel, Executive Director and Chief Executive Officer, ISDA. “ISDA and its members are vigilant to public concerns around transparency. This is yet another measure of increased standardization in CDS.”

This is interesting, particularly given the politics that are swirling around the market. Making the code available to the public could be a first step towards listing some benchmark names on an exchange. As yet, though, the details of the open-source licence are not available.

Hat-tips: Alea via Dealbreaker.