Wednesday, September 24, 2014

2013 HMDA data out; MBS holdings in flux; Student debt impact quantified





I like to put my money to work. I could earn some decent dividend returns (3-4%) if I buy some Exxon, or AT&T. Or maybe I should start loaning my money out like this lender, and earn a 200-300% APR. (Pick a state and slide the dollar amount bar, and watch the APR magically change.) And we wonder why lenders have a bad reputation...or maybe it is merely a matter of risk versus return.

Chase breathed a sigh of relief Monday when, "Virginia Attorney General Mark R. Herring (D) dropped JPMorgan Chase from a mortgage securities lawsuit against the country's biggest banks, after learning that his predecessor Ken Cuccinelli (R) had already struck a confidential settlement with the bank." Huh? Confidential settlements? What about transparency? The plot thickens... 

Who is John Burns Consulting? Aside from working for home builders, I don't know exactly who it is. But its researchers believe that student debt caused 414,000 houses not to be sold. (And I guess the 414k, through the multiplier effect, quickly turns into billions.) But student debt is not going away, although students and families hope for college tuition appreciation to slow down given that the inflation rate is nearly 0%. Millennial Martine Torres suggests, "The real estate industry is up in arms about first time home buyers being saddled by debt and are therefore not purchasing homes. Yes, these factors contribute to the lack of first time home buyers but the mentality of the 'millennial' generation has  shifted from generations in the past that were eager to get married and buy a home to being more career-driven and being content with renting. My generation does not want to get tied up in mortgages at a young age and is delaying marriage; parental dependence is also at an all-time high.  The 'millennial' generation is commitment-phobic. 

Yay! The new HMDA data is out, the new HMDA data is out! The CFPB released the 2013 HMDA data. Remember that the regulators use HMDA data in their examinations to determine whether a lender is complying with fair lending laws. With the addition of more data points in the proposed HMDA regulations, many believe future HMDA data will become more critical in fair lending examinations. For further information regarding the HMDA data, click on the hyperlinks and URLs in the CFPB press release. "The Federal Financial Institutions Examination Council (FFIEC) announced the availability of data on mortgage lending transactions at 7,190 U.S. financial institutions covered by the Home Mortgage Disclosure Act (HMDA). Covered institutions include banks, savings associations, credit unions, and mortgage companies. The HMDA data made available today cover 2013 lending activity, and include applications, originations, purchases and sales of loans, denials, and other actions related to applications."

 

Yes, the HMDA data comes from lenders, and let's check in with some relatively recent announcements to see what some have been up to lately.

 

Per Bulletin 2014-46, issued August 12, 2014, U. S. Bank Home Mortgage announced that "effective immediately for all new FHA applications and FHA loans in process, USBHM was removing the additional reserve requirement on FHA loans when a borrower is vacating the primary residence and converting it to an investment property. The current requirement of 6 months PITIA reserves for both properties is being reduced to follow the current FHA requirement of 3 months PITIA reserves for both properties. Correction: Effective immediately for all new FHA applications and FHA loans in process, USBHM is removing our reserve requirement on FHA loans when a borrower is vacating the primary residence and converting it to an investment property. The current requirement of 6 months PITIA reserves for both properties is being eliminated. No reserves will be required.

A federal court in North Carolina ruled against the Federal Deposit Insurance Corp.'s takeover and multimillion-dollar loan loss claim against Cooperative Bank. The fact that the court ruled against the FDIC's $40 million claim against the former NC bank officers is viewed as a victory for "the small guy." 

Utilizing the Approved eSign Vendors list from Mountain West Financial, electronically signed initial disclosure packages for Conventional, FHA, and VA transactions are acceptable. Investor restrictions do not allow electronic signatures on the Final 1003/Uniform Residential Loan Application (URLA). Loan Originators must "Wet Sign" the Final 1003, no exceptions. 

Franklin American Mortgage's recent bulletin includes updates on Conventional Property Insurance, Private Road Maintenance Agreements, Secondary Financing - LP, DU Release Notes, VA New Construction, Closing Costs, USDA Annual Fee, All Products Incidental Cash Back - Texas, Clarifications FHA & VA Funds Reimbursed Paid by Credit Card, Conventional, VA and USDA Funds to Close Access Letters.

 

First Community Mortgage Wholesale posted guideline changes effective August 31st. These changes include DU Refi Plus and LP RR, FHA, USDA, and Conventional program updates. 

Mountain West Financial Wholesale matrix changes and updates include: USDA limited to 3% Platinum and Sapphire Grant, Mortgage Insurance required on HomePath loans, and removal of Second Homes as eligible for FHA Streamline Refinance. 

New Leaf Wholesale has reduced LPMI rates. The reduced rates will benefit NewLeaf's LPMI offering as MGIC has introduced additional credit buckets to deepen the discount on higher credit scores. New credit buckets are as follows: 680 - 719 un-changed, 720 - 739 lowered, 740 - 759 deeper discount, 760+ deepest discount. USDA loans are subject to the increased guarantee fee regardless of when a loan was submitted for review. If the RD Conditional Commitment is issued on or after October 1, 2014, it will be subject to the higher guarantee fee. No exceptions to this policy are possible. Initial disclosures must reflect the new fee effective immediately.

 

By now, many know that bank holdings of agency MBS has been decreasing. This is an important statistic, once which gives some insight into bank views on prepayment risk, cash flow valuations, and general market risk. As I received a few emails over the past week inquiring if I had any data on the subject, I thought I could combine a few reports I have read. The National Information Center has released consolidated financial statements for bank holding companies, and although the data is not as comprehensive as the Quarterly Banking Profile (which also includes savings institutions), soon to be released by the FDIC, they provide a good early estimate of changes in bank assets and liabilities. What we find is: agency MBS holdings decreased by $4.0 billion for the top 50 banks; a majority of the decline came in conventional agency MBS holdings, which fell by $4.2 billion; agency collateralized mortgage obligations (CMO's, which can be viewed as MBS' eccentric younger brother) holdings decreased $900 million, while the GNMA holdings increased by $1.1 billion; U.S. Bancorp had the largest increase in agency MBS holdings, adding $4.0 billion, while Bank of America reduced its holdings by $8.8 billion; holdings of non-agency MBS of the top 50 banks decreased by $4.4 billion, while commercial mortgage backed securities CMBS (which is MBS' over-achieving cousin who went to an Ivey League school and reminds everyone about it at Christmas time) holdings rose by $7.4 billion; and treasury holdings increased by $48 billion

"There might be a little noise in the data." That, and, "I couldn't figure out how to make a pivot table," are two things you never want to hear coming from your Capital Markets department. However, often times there is a lot of noise in data sets-if there wasn't, then I guess my first boss was correct: you really could turn a monkey into a mortgage trader. Part of the challenge of being an analyst is identifying variables which alter financial models; the good ones make the necessary adjustments, the bad ones ask if you if you all the veggies on your foot long sandwich. BAML writes in Home Prices: Noise Reduction, "We had been using the S&P Case Shiller national composite, released quarterly, in our models. However, this series has been discontinued and replaced with a monthly aggregate, which has a different history. It shows more stable prices, with less of a decline during the housing bust and a more steady recovery since bottoming at the end of 2011. According to this new national composite, home prices only fell 26% peak-to-trough (versus the prior estimate of 34%) and have increased 19% from then." The consequences of the new composite, according to the article, is the revised data suggests home prices will be up 3.9% this year (4Q/4Q) compared to BAML's prior forecast of 5.1%, and 3.4% next year.  

Things are heating up in the pipeline hedging biz. Compass Analytics, LLC announced that it is opening a new office in Midtown Manhattan on Lexington Avenue. "The new office is 2 blocks from Grand Central Station and provides easy access to all neighborhoods in New York City and the surrounding areas.  This Midtown Manhattan office allows us to continue to add to our presence on the east coast, making it more convenient for our customers to visit with their account managers and to also allow for more regular visits to many our Clients based in the North East."

 

For the bond markets, lack of volatility continues to be the case although by the end of Tuesday the 30-yr agency MBS "stacks" finished about .125 higher and the 10-yr risk-free U.S. T-note closed at a yield of 2.53%. So far this morning we've had the MBA's application numbers (not the same as locks) for last week. They dropped over 4% with purchases down slightly but refis down 7%. And at 8AM MST (MST since I'm in Denver today) will be New Home Sales for August; it is seen gently improved versus the prior print (+412k).  In the early going we're roughly unchanged at 2.54% on the 10-year and agency MBS prices better by a smidge.

 
http://globalhomefinance.blogspot.com

1 comment:

  1. Low- and middle-income students worried about the consequences of taking out a loan will be more likely to decide that college attendance is not worth the risk as denoted by Epic research.

    ReplyDelete