As we all know, we had a credit and real estate bubble that first lifted then toppled nearly all of the major US banks. Apparently, none of the gurus on  the Street saw this coming (the very same gurus who now say we are in recovery and the worst is behind us), save a tiny coterie of truth seekers (Did Reggie Middleton, a Blogger at BoomBustBlog, Best Wall Streets Best of the Best?). The downfall of the banks was basically binging on inflated, junk assets using excessive leverage at the top of a big bubble. The government used the alphabet soup of programs to bailout the banks, but the mainstream media focused primarily on TARP of about $750 billion, which kept everybody's focus off of the real money that ran well into the tens of trillions (listed in the 3rd paragraph above). The government then agreed to buy many of the aforementioned junk assets off of the banks using programs that I warned were truly suspect. Now that we have caught up to recent history, let's move on...

These mortgage and real estate related assets are of very little value to anyone outside of  vultures and speculators looking to purchase them at a deep discount, despite the fact that the government has overpaid for them continuously with taxpayer monies. Bank of America is trying to sell $1 billion of toxic mortgage assets, and these assets have already been written off, which goes to show you have bad they want to get rid of them vs trying to work them out to achieve maximum recovery. Maybe the bank has come to the same realization that I have been espousing for years, that there is little to no effective economic recovery to be had.

About a month ago I explained that "The Robo-Signing Mess Is Just the Tip of the Iceberg, Mortgage Putbacks Will Be the Harbinger of the Collapse of Big Banks that Will Dwarf 2008!" and literally days later Bank of America was hit with a massive $47 billion put back request from investors. The CEO vowed to fight the requests, but now seems amenable to settle. What do you think all of the other investors will do if or when this bank (or other banks) cave without a fight? Now we have Bank of America negotiating with PIMCO over $47 billion of put backs:

Bank of America Corp., after vowing to fight requests that it repurchase certain loans, has begun potential settlement discussions with some of its largest mortgage investors, according to people familiar with the situation.

The group discussing a possible settlement with the nation's largest bank as measured by assets includes the Federal Reserve Bank of New York, government-owned mortgage company Freddie Mac, BlackRock Inc. and Allianz SE's Pacific Investment Management Co., or Pimco, a unit of Allianz SE.

The approach appears to be a major shift in strategy for Chief Executive Officer Brian Moynihan, who in November pledged to engage in "day-to-day, hand-to-hand combat" on investor requests to repurchase flawed mortgages made before the U.S. housing collapse.

Now, what is really the value of these securities and assets that Moynihan is trying to dump so aggressively? Let's look at the Fed's recent disclosures to find out. The asset buying binge that we are focusing on today was aided by the Public/Private Investment Program (PPIP, see Reggie Middleton’s Overview of the Public-Private Investment Program). The “stated goal” of the Public-Private Investment Program is "to strengthen capital base of financial institutions and enhance their ability to lend, ensure efficient price discovery of legacy assets by involving private players and minimizing the risk to taxpayers while providing opportunity to private players to earn sufficient returns." I have raised concerns that this program is actually a backdoor bailout and would enable sole sourcing in lieu of open market, competitive bids. Thanks to Bloomberg’s FOIL suit against the Federal Reserve, we can now see how correct I have been in my warnings and proclamations regarding funny business in the MBS markets. Let’s take a look see.

The Government’s MBS Purchase Program

As we highlighted in Reggie Middleton’s Overview of the Public-Private Investment Program, price discovery has to occur at a natural level for a sustainable bull market which was not a prominent feature of the PPIP. In “Reggie Middleton on PPIP, part 2” Thursday, March 26th, 2009 I gave explicit examples of how collusion and price manipulation could take place in the absence of true price discovery. Fast forward to today, and we find that the Fed’s newly minted MBS sales disclosure page states that “Outright [MBS] purchases were conducted via competitive bidding to ensure that trades were executed at market rates.” However, according to a paper (h/t EB from ZH) entitled “Large-Scale Asset Purchases by the Federal Reserve / Did They Work?“, written in part by NY Fed SOMA Manager, Brian Sack states that “Because the MBS purchases were arranged with primary dealer counterparties directly, there was no auction mechanism to provide a measure of market supply. Instead, the pace of purchases of each class of MBS was adjusted in response to measures of whether that class appeared relatively cheap or expensive. To avoid buying at excessively high prices and to support market functioning, purchases were increased when market liquidity was good and were reduced when liquidity was poor.” Sounds like somebody’s stretching the truth a little, doesn’t it? Well, I dug a little farther to see just who that somebody may have been.

The truth swept behind the 340,000 data item spreadsheet, buried beneath over 70,000 transactions and over 10,000 MBS-only transactions reveals....

We have conducted analysis on all MBS sale and purchase transactions conducted by the Fed whose data was recently released. Of the total 10,058 MBS transactions, 72% were done at a yield of less than 5% (5% below yield of 4.0%, 32% between 4.0%-4.5%, 35% between 4.5-5.0%) with an average yield of 4.75% on all MBS transaction. The table below presents the number of transactions under their respective yield category.

We have also analyzed the yield on MBS purchased and MBS sold, looking for price discrepancies between MBS purchased and MBS sold. The data points out that the average yield on MBS purchased was 4.71%, 29bps lower than average yield for MBS sold, thus implying MBS purchased were at a higher price than MBS sold. You know that old government adage, buy high and sell low!

Yield on sale: 5.00%
Yield on purchase: 4.71%
Difference in bps: 29.1

Assuming a 4.0% implied yield on all securities, 95% of the MBS securities were purchased at a premium to market value while assuming 5.0% implied yield 28% of securities would have been purchased at a premium to market value. Of course, the question remains… Why pay a premium to market value at all (with even .01% of total purchases) in a distressed and downtrending market with highly questionable collateral? Had the government/Central Bank followed the prudent man rule and paid a slightly higher yield (avg yield of 5.0% instead of 4.75% - basically a discount for the assets as is called of in distressed buying), it would have saved $62bn of tax payers’ money on MBS transaction while a 6.0% and 8.0% yield would have saved $391bn and $869bn of tax payers’ money, respectively. Please keep in mind that Ex-secretary Paulson’s initial TARP request was for a mere $750 billion. One could be rest assured that the private sector using its own money at full recourse will be looking for steep discounts, unfortunately our fair government was all too generous.

Other observations

Deutsche Bank with $411bn of trade volume, had the highest MBS transaction value followed by Credit Suisse at $383bn and Morgan Stanley at $280bn while Canto (4.55%), UBS (4.60%), Nomura (4.61%) had more favorable yields compared with other banks (average yield for transaction was 4.75%) The table below presents information on MBS transactions and yield for each participating financial institution.

All paying subscribers should feel free to download the scrubbed and analyzed data for all banks, primary dealers and investment managers here: Federal Reserve MBS Purchasing Analysis. I have made it easy to go through the data by sorting through the 340,000 or so data points for you and putting them in a neat little Excel model. Those who are interested in subscribing should click here.