Rate Watch #757 Mortgage Securitization
January 14, 2011
by Dick Lepre
dicklepre@rpm-mtg.com
www.loanmine.com
Analysis
I keep reading discussion about recovery in the housing market. Supply will stay high as foreclosures are pushed into the future and these discussion miss the point. Until we have some resolution to the demise of FNMA and FHLMC and a new machine for securitizing mortgage debt, the future of the housing market will be uncertain. The questions are these: 1) is the government going to back new mortgage debt and 2) if it does not what is the increased risk premium (rates) which will be demanded by investors? If Fed monetary policy fuels inflation then higher rates will hurt housing. Jumbo mortgage securitization is starting to return.
A weak housing market hurts GDP. There are 2,000,000 fewer people in construction than there were at the height of the housing bubble. There is unlikely to be significant uptick in construction until 2013 at earliest.
FNMA will keep the current conforming and high-balance conforming loan limits in place for loans funded before September 30, 2011. After that, the high-balance conforming will be cut back (as word fail me here) to the permanent temporary limit of $625,500.
A list of the high-balance limits by county may be found here by selecting the state from the pull-down and typing the county.
Mortgage Securitization
Let's consider three things about mortgage securitization:
1) why does it exist and what are its benefits?
2) how did mortgage securitization become one of the biggest causes of the mortgage mess/Great Recession?
3) what should we do in the future?
Why Does Mortgage Securitization Exist?
Think about banks. Banks have deposits which are almost entirely short term. These are checking and savings accounts and CDs (certificates of deposit.) If inflation were never an issue and interest rates on savings and loans never changed we would have less need for mortgage securitization. Mortgage securitization exists because inflation exists and the cost of funds to banks changes. If a bank cannot predict what its cost of funds will be in the future there are only three possible approaches a bank can take to mortgages: 1) do not do them at all 2) do only adjustable rate mortgages or hybrids such as 5/1 ARMs or 3) transfer the rate risk to another entity.
Assuming that banks will make mortgage loans and assuming that borrowers want the safety of fixed rates then mortgage securitization is a necessity. Mortgage securitization allows banks to fund mortgage loans and sell them at a fixed price to investors who will keep them long term. These investors are wealth holders. These should be entities which are prepared to hold these securities long-term.. In this system, everyone acts and an intermediary between the borrower and the investor.
The biggest entities in mortgage securitization both before and after the mortgage mess were FNMA and FHLMC. What went wrong with these?
President Clinton saw declining or flat homeownership rates as problematic and wanted to increase the rate of ownership. He wrote this to HUD Secretary Henry Cisneros on November 3, 1994: "Today, I am requesting that you lead an effort to dramatically increase homeownership in our nation over the next six years. I would like you to work with the Assistant to the President for Economic Policy, the Assistant to the President for Domestic Policy, the Secretary of Agriculture, the Secretary of Veterans Affairs, and other private and public sector partners you may designate to develop a National Homeownership Strategy that will carry us into the 21st century."
What was the result? Let's read from a a document, called "The National Homeownership Strategy: Partners in the American Dream," created by HUD in 1994: [boldface added]
"For many potential homebuyers, the lack of cash available to accumulate the required downpayment and closing costs is the major impediment to purchasing a home. Other households do not have sufficient available income to make the monthly payments on mortgages financed at market interest rates for standard loan terms. Financing strategies, fueled by the creativity and resources of the private and public sectors, should address both of these financial barriers to homeownership."
Some of the thinking behind this remains on the HUD web site at http://www.huduser.org/publications/txt/hdbrf2.txt. HUD's thinking was motivated by progressive social concerns. Folks asked "Why the heck should the government be implicitely backing FNMA and FHLMC if they only do loans for folks with good credit and good debt ratios? No one saw the consequences of doing otherwise. HUD reveled in the success of the homeowenership strategy and President George W. Bush talked about how great it was that we had record homeownership rates. Bush thought this was so great that he signed the American Dream Downpayment Act. In the words of HUD, "ADDI will help first-time homebuyers with the biggest hurdle to homeownership: downpayment and closing costs. The program was created to assist low-income first-time homebuyers in purchasing single-family homes by providing funds for downpayment, closing costs, and rehabilitation carried out in conjunction with the assisted home purchase." The housing bubble now had bipartisan support.
There were two problems with The National Homeownership Strategy: 1) making loans to people who had little down payment and who did not have sufficient available income to to make the monthly payments should never have gotten outside the relatively strict documentation standards of FHA which was a long-standing government guaranteed program for precisely this purpose. Insisting (as HUD later did) that FNMA and FHLMC make more their half of their new loans to people who would not previously have gotten FNMA/FHLMC loans was the largest, by far, reason for their demise. 2) by creating so many more folks who could now buy a home at a low interest rate a housing bubble was created and the Great Recession was an inevitable consequence.
HUD may have started the bubble but it had a lot of help. About 41% of what had perviously been subprime mortgages were held by FNMA and FHLMC. The rest were sold into PLMBS (private label mortgage backed securities). The biggest player here was Countrywide which, because of the massive size of what it was able to produce, had the ability to go to Wall Street to securitize there. In my view the failure here was with the debt rating firms (Moody's, Standard & Poor's, and Fitch) which enabled this bad debt to be sold as investment grade securities. There was a regulatory failure here in making these the firms an oligarcy. Since they had no competition they had less incentive to do the job correctly.
But We Were Talking About Mortgage Securitization
Right. My point was that is was not mortgage securitization that was the problem but that the securitization of bad mortgages expanded the size of the housing bubble.
One of the ongoing problems I have with the reporting (and this has been going on for > 4 years) is that this is still presented as "banks gone wild" as if banks made all these crappy mortgages and foisted them off on others. The post-Lehman liquidity crisis existed entirely because banks held mortgages not yet delivered into the system and, worse yet , substantial amounts of mortgage backed securities. One of the problems was banking regulations themselves. The Basel accords are an international standards for specifying how banks should count reserves. Basel I (the accord in effect in the U.S.) specifies that an adequately capitalized bank should have capital equal to no less than 8% of its assets. Remember that a bank's assets are its loans and its liabilities are its deposits. But Basel recognizes that different assets have different risks. For example, Treasury debt has zero risk. A commercial mortgage has 100% risk. A $100 loan to a business ties up $8 of a banks' capital. A $100 loan to the Treasury Department ties up $0 of its capital. A mortgage has a 50% risk so a $100 mortgage loan ties up $4.00 in capital. BUT mortgage backed securities have a 20% risk so that $100 worth of MBS ties up $1.60 of capital. Converting whole loans to MBS reduced the bank's need for capital by 60%.
So what did banks do? They sold the mortgages which they made (and presumably knew more about) and purchased 2.5 times as much face value MBS with their capital. The effect was that they had a much larger exposure and went broke faster. The street wisdom is that the bank executives made a ton of money. The fact is that almost all large banks (the exceptions being J.P Morgan Chase and Wells Fargo) wound up holding a ton of mortgage debt and it was the decrease in the value of that debt which led to the liquidity crisis.
Further push down the path to hell came from SEC - another government regulatory entity. In 1999 SunTrust Bank was penalized by SEC for keeping "excess" loan loss reserves. It was the contention of SEC that SunTrust was doing this to make earnings look less volatile. SEC's reasoning was "there is no possible way that you are going to have losses like that." Maybe SEC was correct and SunTrust was smoothing their earnings but what we have learned was simply that very few banks had enough reserves when the bubble burst. SEC's ruling discouraging increased reserves was not their best idea ever. In fact both the Federal Reserve and later FASB (Financial Accounting Standards Board) sided with SEC.
The seeming advantage of leveraging capital by purchasing more mortgage debt combined with the diminished lending standards of FNMA and FHLMC mandated by HUD under the National Homeownership Strategy worked to create the housing bubble which, when it burst, created a bank liquidity/solvency crisis.
But We Were Talking About Mortgage Securitization
OK OK. My point was that there was and is nothing inherently wrong with mortgage securitization. We should learn several things from the mortgage mess: 1) insisting (as HUD did) that FNMA and FHLMC lower their lending standards without readjusting the Basel capital requirements was a disaster. These loans were riskier than loans which the GSEs had made previously but no one was paying attention to that.
There are obvious solutions: return the lending standards of the GSEs back to the way they were before The National Homeownership Strategy. This has already been done. Make more realistic loan level price adjustments (adjustments to price for LTV, credit score and other risk factors such as cash out.) This has been happening since the start of 2009 and was just changed again this past week. Now, in order to get the best pricing you need a 740 middle credit score and an LTV of less than or equal to 70%. It appears to be the case that loan prices are becoming sensitive to LTV. The add-on for LTV of 80% with 740 credit score is small (0.25 points in fee.)
I believe that by far the biggest single factor which would make sure that loan payments are made on time and that the cash flow of MBS is preserved is the loan to value. If someone has little down payment there is not sufficient motivation to find a way to make the money to make the payment. If I have $100,000 of my own hard earned money in my house in the form of down payment then I may be willing to find a way to earn money to make the payment. If I have no "skin in the game" or worse yet, my mortgage is $100,000 more than the value of my home how much motivation is there to find a way to make the payment?
Mortgage securitization done correctly benefits the economy but, like Longfellow's girl with the curl in her hair, when mortgage securitization is good it is very, very good. But when it is bad, it is horrid. Mortgage securitization done incorrectly greatly amplified the underlying mistakes which created the housing bubble. Let me be clear: there was more subprime, Alt-A lending done outside of the GSEs. Private label subprime was done in massive quantities by Countrywide and National City. These were securitized as investment grade paper with the imprimatur of the three debt rating firms.
Worse yet, whatever the underlying problems may have been, the aggressive monetary policies of the Federal Reserve poured gasoline of the fire.
The fact is that before the mortgage mess we absolutely knew how to do mortgage lending and mortgage securitization. The reason why we chose to do bad loans and mandate securitization of bad loans are in the past. It is not important to me who is to blame. It is important to know that we know how to do good mortgages and good mortgage securitization and we can do so going forward.
So what do good mortgages look like? They are fixed rate and fully documented (or, at present, over-documented). The borrowers have debt ratios of no greater than 45% and evidence of income stability. They demand 20% down payment and give the best price for 30% down payment. They make appropriate adjustments for risk-factors such as 1) investment property 2) property type and 3) declining markets.
There has been a lot of speculation about what is going to happen with FNMA and FHLMC. In the short run nothing is going to happen. Eliminating these and the Treasury guarantee at present would be a disaster. The loans made in the past couple of years are of substantially higher quality than those made during the bubble. The troubles that these GSEs had are in the past and were consequent to The National Homeownership Strategy. In time, when there is confidence that real estate values are no longer falling I see three possibilities: 1) these entities can be reconstituted as they were as corporations whose debt is backed by Treasury 2) these entities can be reconstituted as corporations whose debt is NOT backed by Treasury 3) securitization can be taken over by investment banks and the GSEs eliminated. In fact if I worked for an investment bank I might be interested in starting a completely private competitor to FNMA to securitize mortgages. The natural place to start is with jumbo loans because there is no competition from FNMA, FHLMC, FHA or VA. At present, we are starting to see the reemergence of jumbo mortgage securitization.
There should be regulatory adjustments. Banks were allowed to move loan portfolios "off sheet" by creating OBSEs (Off Balance Sheet Entities) such as SIVs (Structured Investment Vehicles). These SIVs were set up to expire and roll over each year. The fact that these had duration of less than one year enabled these SIVs to complete escape the Basel capital requirements because Basel assigns zero risk to short term exposures. Bad idea.
In addition, we need to recognize that at some time in the future there will be another housing bubble and that it serve well to ratchet up capital requirements for whole mortgages and MBS at any time in the future when housing prices are rapidly inflating. The fact is that when the bubble was in full swing, everyone was less concerned about good mortgage underwritng because there were "outs" for folks who could not afford their payments: they could sell or the could refinance. Once values started down in 2006 the collapse was inevitable. We need to note for a future generations that when housing prices start moving up by 10% a year for several years it is time to tighten lending standards not loosen them.
Dick Lepre
RPM - SF
1400 Van Ness Avenue
San Francisco, CA 94109
DRE License # 01143973
NMLS Individual ID 302379
dicklepre@rpm-mtg.com
Web site: www.loanmine.com
Blog: economy.typepad.com
(415) 244-9383
(866) 488-2051 fax
California Department of Real Estate - real estate broker license #01201643
Recent Comments