Flagstar and Castle & Cooke; MBS Margin Requirements Could Hurt Locks, Pricing and Borrowers
In this video, Wells Fargo voices its stance on the upcoming changes.
The CFPB gave renewed reason for many lenders to contemplate taking their chips from 2012 and 2013 off the table and contemplate another occupation. (And if lender ranks drop dramatically, is the borrower better off?) "The Consumer Financial Protection Bureau (CFPB) announced a proposed consent order in its enforcement action against Castle & Cooke Mortgage, LLC, for allegedly steering consumers into costlier mortgages. The Bureau has asked a federal district court to approve a consent order that would provide more than $9 million in restitution for consumers and obtain $4 million in civil money penalties against Castle & Cooke and two of its officers for allegedly paying loan officers illegal bonuses." Here is what the public saw.
A $13 million fine against a company doing about $100 million a month is really a head-turner. The resolution of the lawsuit that the CFPB filed against Castle & Cooke Mortgage and its officers in August appears to be exactly that - a resolution. A consent order reflecting the settlement was filed in federal, and under the order the company and its officers admitted no wrongdoing. It is believed that senior management knew of the monetary figure two months ago and agreed to it rather than spend money fighting it, allocating the money at that time. And according to inside sources, "It is full speed ahead here at C&C, and we're going to continue with our track record over the last several years of greater than 80% purchase business. And in fact, if one takes the time to look at the proposed consent decree (found in a link near the bottom of this page) it is indicative of a company that cooperated with the CFPB in this matter.
Five states over in Michigan, Flagstar Bancorp, the holding company for Flagstar Bank, FSB, announced that it has entered into an agreement with Fannie Mae to resolve repurchase requests and obligations associated with loans originated between January 1, 2000 and December 31, 2008, for a total resolution amount of $121.5 million. After paid claim credits and other adjustments, the Bank will pay $93.5 million to Fannie Mae. At September 30, 2013, Flagstar's total representation and warranty reserve was $174.0 million and the amount of the reserve specific to the loans covered by the agreement was sufficient to cover the payment amount. The agreement covers the bulk of the loans originated between January 1, 2000 and December 31, 2008 and sold to Fannie Mae, regardless of whether Fannie Mae has made a repurchase demand on any particular loan to date.
And Fifth Third Bank is not immune from legal action. "The U.S. Justice Department and Department of Housing and Urban Development's Office of Inspector General are investigating Fifth Third's practices related to loans that were insured by the Federal Housing Administration."
I often receive questions about the mortgage-backed securities market, especially the size of it. The recent prepayment speed numbers provide some indication into current production trends. The total agency MBS production for October was $52 billion, down $45 billion (47%) versus last month. Breaking that down, 30-yr production was $38 billion, down $37 billion (49%) versus September while 15-yr production was $10 billion, down $6 billion (37%). Remember, there's an 8-10 week lag between primary mortgage rates rising & homeowners closing on their purchase/refi and the loan being put into a security. But investors watch these numbers, as clearly the refi machine began to shut down starting in May when the 10-yr Treasury started its move toward 3.00% and the 30-yr conventional mortgage rate shot from a low of 3.25% to 4.75%.
The Fed, however, remains a strong technical demand force currently. It is buying nearly $3 billion a day compared to mortgage banker supply which is running between $1-1.5 billion per day (about half of where it was six months ago). The Fed's buying is heavily concentrated in 30-year 3.5s and to a lesser extent in 4s - pretty much mirroring current production. I have spoken to traders who deal with the Fed, and say that their purchase methods aren't particularly complex or sophisticated. Maybe they don't have to be...
I am also occasionally asked what a "swap" is, and why anyone should care. I say that it has nothing to do with your friend's marital practices, and is an important tool for banks that don't necessarily want 30-yr fixed-rate assets stacked up against very short term deposits. I penned "A Primer on Swaps, and the Implications of Change in the Secondary Markets" to give folks some elementary education on them, and why they should care.
"Rob, are you hearing anything regarding the change in margin requirements for TBA's? I have heard rumors. And my Capital Markets guy says that, if true, this is a significant change by Fannie and we will no longer be able to avoid margin calls on our trade lines like we've done in the past - the threshold amounts will determine the impact. What's up?"
Excuse the litany of acronyms, but yes, you are correct, there are some changes being made to the way agency TBAs (To Be Announced securities, the favorite MBS vehicle used to hedge pipelines and protect companies, and borrowers, from interest rate movement) will be margined. The TMPG (Treasury Market Practices Group) has required all primary dealers to incorporate bi-lateral margining in their MSFTA's (Master Securities Forward Transaction Agreement), and has recommended that all parties do the same.
Dan McPheeters, who is the MBA's Policy Advisor, Public Policy & Industry Relations Manager, wrote, "FINRA is embarking on a re-write of Rule 4210 and will tackle, among other things, positions in agency TBAs. I expect their revised rule proposal to come out for comment in early December. We have been actively engaged with FINRA for the past 3+ months on this issue, and they will be speaking on this topic on a panel during MBA's IMB conference in December. FINRA is in the middle of an extensive rewrite of their margin requirements and the early reviews were not good for mortgage bankers. The MBA has built a very good dialogue with FINRA, and they are aware of the problems of an 'all best efforts' world, and are much more familiar now with the business model. So I would say things are looking much better than they were a couple months ago. That said, it takes time and a LOT of money to build out scalable margin exchange and compliance systems, so my guess is much of the chatter is from those who have invested early in these systems. There is also a lot of misunderstanding concerning what rules are going into effect when, so never underestimate the fact that folks just may not know." (If you have questions on current developments, write to Dan at email@example.com.)
And Fannie's Trading Desk mirrored this information in a note to its clients. "In November 2012, the Treasury Market Practices Group ("TMPG"), an industry group sponsored by the New York Federal Reserve, issued a recommendation that industry participants begin exchanging variation margin on forward-settling agency MBS transactions (see http://www.newyorkfed.org/tmpg/margining.html). The TMPG-proposed implementation date for these changes is January 1, 2014. Fannie Mae intends to follow the TMPG's recommendation and implement the ability to exchange bilateral margin with lenders who transact with our Capital Markets Sales Desk. This change will be implemented by a Selling Guide Announcement that will update and amend section C3-7-01 of the Selling Guide. This section of the Selling Guide, which sets forth various requirements and procedures around establishing an MBS trading account with Fannie Mae, will be updated to incorporate these new margin requirements, including pre-set threshold amounts and minimum transfer amounts relating to exchanging variation margin. We intend to make the Selling Guide Announcement in December 2013 in order to have these changes go effective January 1, 2014."
What does all this mean to rate-sheet prices? If the various regulators and regulations further crimp, or make more costly, the hedging activities of residential lenders, this cost will be passed on to borrowers. What if an LO couldn't give a 30-day rate lock to a borrower? So it is indeed important - we wish Dan McPheeters and the MBA success.
"Rob, do you have a simple explanation about rates go down when bond prices go up, and vice versa, and what 'duration' means?" I'll give it a shot. If rates move higher, existing bond prices call since investors will be able to purchase new issues that pay better rates, making existing bonds less attractive and reducing their market value. Everything else being equal, why would anyone pay $100 for a 30-year bond yielding 2.00% if they could pay $100 for a 30-yr bond yielding 2.50%? But they might pay $97 for that same bond yielding 2.00%, thus making the yield 2.50%. (The opposite occurs when interest rates fall and new issues pay lower rates. When this happens, investors are willing to pay higher prices for existing fixed-income securities - like mortgage-backed securities - that pay better rates than new issues.)
Duration has a couple different meanings but try to avoid equating it with "maturity." (Remember the old joke - why are men like bonds? They don't pay much interest and they rarely mature.) To an analyst or investor, duration takes the relationship between interest rates and bond prices a step further by measuring the sensitivity of a bond's price to a change in interest rates. More specifically, duration can be used to estimate how a 1% shift in interest rates may affect a particular bond's market price. The longer (higher) the duration, the more a bond's price should fluctuate as interest rates rise or fall. For example, the price of a bond with a duration of five years would be expected to fall 5% for every 1% increase in market interest rates. Conversely, a bond with a duration of ten years should rise 10% for every 1% decrease in rates. Treasury 10-yr notes might have a duration of 8.80 at a yield of 2.50% versus 30-yr bonds having a duration of 18.50 at a yield of 3.50%. Traders and pipeline hedgers are very focused on duration - long term bonds tend to have longer durations and their prices can fall quickly when interest rates increase. But the impact on MBS prices is influenced not only by that, but also by the likelihood of a borrower to refinance, and/or the value of servicing for that mortgage or pool.
Yesterday we had the GDP news (Gross Domestic Product rose at a 2.8% annualized rate, much stronger than expected) which was balanced out by Consumer Spending (up 1.5%, the smallest increase since 2011). The numbers continued to indicate steady growth but there isn't much in the U.S. economy to get excited about. By the time the dust settled Thursday agency MBS prices were better by about .125 and the 10-yr closed at 2.61%.
But today we've had the October employment numbers. Nonfarm payrolls were expected +125k, but came out at +204k with a back-month revision of +60k. (The unemployment rate stands at 7.3%.) In addition, we had October Personal Income (expected +0.3 versus +0.4 last) and Consumption (expected +0.2 versus +0.3). The very strong employment numbers has the fixed-income market back on its heels. The 10-yr closed Thursday at 2.61% and today it is at 2.73% and agency MBS prices are worse by 1 point! And hey, don't forget that the bond market is closed Monday for Veteran's Day - so even for the investors that might be open, don't expect cutting-edge pricing.