Retired REO broker Ramsey Su has sent out another set of observations on the housing market, this time covering three topics: the "subprime" spillover, the imbalance between housing supply and demand, and an upcoming huge foreclosure auction here in San Diego.
Subprime Spillover, Supply and Demand, and Foreclosure Auctions
(Ramsey’s disclaimer: this email is unedited; take it for whatever it’s worth)
Definition – there is little doubt now that the term “subprime” has been misunderstood and misused. "Non-traditional" is a much better term. High debt-to-income (DTI) ratios, 100% combined loan to value (CLTV) ratios and loans with multiple layers of risk are far more accurate descriptions of the problems that haunted the market of late. I am going to use “non-traditional” in lieu of what is generally labeled “subprime”. Similarly, labeling the “prime” loans as “traditional” loans would provide a much better understanding of the spillover effect.
Spillover before 2007 – traditional loan delinquencies had no logical reason to occur. Borrowers with traditional loans, when in financial trouble, had two simple alternatives. The best alternative was to sell the property. The second alterative was to simply convert themselves into non-traditional borrowers via a refinance. In either case, anyone who concluded that since the delinquency rate of traditional mortgages have not gone up as rapidly as the non-traditionals, and therefore there was no spillover effect is most likely mistaken.
Fitch has some numbers that support my points above:
"While FICO score continues to be a highly predictive measure of relative credit risk among loans with similar characteristics, layering on additional risk factors causes absolute default rates to be higher." — Fitch
Spillover 2007 and beyond – the major event at the beginning of 2007 is the removal of non-traditional mortgages from the market. Compounded by lack of price appreciation, many traditional borrowers in trouble have no intermediary step. If market conditions do not stabilize, we should soon see defaults in the traditional loans. If this happens, it would be a great indicator that the real estate bubble is far more damaging than a soft landing.
NAHB – David Sieders recently estimated that – A record volume of vacant (unoccupied) housing units is weighing heavily on both single-family and multifamily housing markets, and historical norms suggest an excess of about 1.3 million vacant units on the market.
Census Bureau – Reports on Residential Vacancies and Homeownership: http://www.census.gov/hhes/www/housing/hvs/qtr107/q107press.pdf
The Census Bureau is showing record vacancies and declining percentage of homeowners.
Homebuilders’ earnings – so far this quarter, it is unclear that any builder has curtailed their building activities. While most claim they are starting fewer new specs, almost all have added communities. At moments notice, they can all be building and building. Furthermore, when builders reduce land inventory, it is not as if the land vaporized. It is ownership that temporarily changed, or reverted back to the original seller when options are not exercise. The supply of land remains unchanged.
In summary, supply is rapidly increasing via foreclosures and new homes. These auctions (see below) are going to show how much pent up demand there may be on the sidelines and what price level is necessary to bring supply and demand back to equilibrium. If these auctions prove to be successful, lenders and builders alike are going to repeat the same strategy until the market turns or auctions are no longer returning the desired results.
As a real estate investor, what would I pay for these properties? It was an easy decision; I would not touch these properties. If owner users show up in volume, then I do not want to compete. If participation is weak, then it is clearly a falling knife.
Both auctions will be completed in two weeks or less. We will find out soon enough.
This is a big auction for so early in the REO cycle. Well worth watching.
This appears to be Lennar’s attempt at a quick liquidation of inventory in the desert before it gets too hot.
I did some forensic work on about half of the 96 properties scheduled to be auctioned in San Diego on May 12th. It was thought provoking, affirming some previous theories and formulating some good leading indicators for future trends.
The following are miscellaneous notes I made while analyzing the auctions. Some of these subjects should receive more future attention.
Profiling the original loans – all the loans were originated during 2004, 05 or 06. None of the REOs were results of a junior lien, all were 1st lien foreclosures. In other words, all the “20s” of the 80/20s were wiped out. In my opinion, the dominance of the 80/20 REOs is a confirmation that this is a financing (reset) triggered foreclosure cycle, and not a result of traditional financial hardship such as job loss or other unexpected increase in expenses. This is consistent with my previous studies of SD REOs.
Beneficiary’s bid at trustee’s sale – if there are bidders at the trustee’s sale, the beneficiary (lender) will bid their encumbrance plus all delinquencies and expenses. If there are no bidders, the lender typically bid a lower amount. An accurate bid at sale figure would have been very useful because that tells me the book value of this particular asset from which we can calculate subsequent losses. So far, most of the bids at sale are below the encumbrance, not to mention other costs. Furthermore, there is no consistency from lender to lender. It may be better to use the original loan amount plus some cost factor to use as the lenders original basis.
Original loan amount (1st) – is a very meaningful number, more so than the bid at trustee’s sale which may be quite arbitrary. As an example of one of its uses, in previous emails, I have shown that the hard cost of the REO disposition process is about 25%. So if the auction price is exactly the loan amount, we can assume that the holder of the note lost 25% on this loan. (Not including any junior liens which had already been wiped out)
Last list price (LLP) – is also a very meaningful number. Many of the auction properties had been listed with local REO brokers before being pulled for the auction. The lenders/servicers typically determine the LLP using a combination of an appraisal and a broker’s price opinion (BPO), plus a few percent for some negotiating room.
Quick Sale Value (QS) – is a crude but probably very accurate method of coming up with a bench mark to judge the success or failure of this auction. I take the properties that had been listed which would have an LLP, then remove 5% as the fluff in the asking price, then assume a 10% discount would be adequate to sell properties under the current market conditions without resorting to an auction. Now all I have to do is watch the auctions results and look at the over/under percentages of each of the final bids.
“Opening bid” (OB) – this is a meaningless number since it is not the reserve. However, lacking the actual reserve, I have to use something as a starting point for analysis. Since there is a 5% buyers’ premium, I am using the published opening bid +5% as the true opening bid.
“Previously valued to” – this value, when available, appears to be the last sale or last refinanced amount. It is as meaningless as the opening bid. If the property is worth remotely close to this value, there would be no REO and no auction.
Buyer’s premium – buyers pay bid price plus 5% premium as final price. While this is common in art auctions, I don’t remember seeing that used in real estate auctions. I can only assume the auctioneer would repeatedly remind bidders of this premium or else there would be surprises in the closing room.
Financing – Impac is the chosen lender. I was told the maximum LTV is 95% regardless of owner or non-owner occupancy. The other terms are basically whatever Impac has to offer based on the bidders’ qualifications. The only incentive is a $500 reduction in closing cost with Impac. Other lenders are acceptable but limited to CFC, GMAC, WFC, WM or BAC. Though underwriting standards are far more stringent today, I do not see financing as an obstacle. However, the pre-qualification process would bring affordability issues to the forefront. This auction may tell us how many truly qualified buyers are sitting on the sidelines waiting for a good deal, or have they already all bought during the last few years.
LLP as percentage of 1st – this gives us a ballpark figure of how much property value might have depreciated since the origination of the loan. I have a sample of 16 properties and the average is 118%. Since these 1st liens were 80% of the original sales price, this would suggest a 2-3% depreciation IF the properties can be sold at the LLP.
QS as percentage of 1st – this gives us a likely sales price under current condition. For the same 16 properties, this percentage is 100%. This would imply that the properties have depreciated 20% (the junior lien amount) and the 1st is going to take a 25% loss on this loan, in other words, the cost of the foreclosure.
OB as percentage of QS – this comes out to about 67%. While this is a meaningless number, I believe it is low enough as a teaser to draw the attention of potential buyers in the market today.
How I would grade the Lennar auction – if LEN manages to sell all the properties, I would consider it a success. Now if they would stop building more than they can sell, that would really be a success. If not, there will be more and more of these:
How I would grade the results of the REDC REO auction:
A – >110% of QS value (net of 5% buyers’ premium)
B – 105% to 110% of QS value (net of 5% buyers’ premium)
C – 100% of QS value (net of 5% buyers’ premium)
F – <100% of QS value (net of 5% buyers’ premium)
I expect the results to be an “A”. Anything less than that would be quite negative for the market. Here are the negative factors:
Very expensive – auctions can easily cost double that of a regular broker’s commission of 5-6%.
Alienate local brokers – regardless of what the opinion of local brokers may be, they are responsible for the majority of the sales. Any broker who has a good, qualified buyer today is sitting on a solid 3% commission. Why would brokers with good buyers take their valuable clients to an auction for 1% when they can go to a new homes community and get the moon? Furthermore, 99.9% of brokers have never represented a client in an auction, nor are they familiar with how auctions work. They have no reason to take on additional risk for less pay. They will all be trying to steer their clients away.
Kills the market – every buyer for similar properties will either try to “steal” at the auction or will choose to wait for results. Just imagine the legit buyers driving around the neighborhood that they are shopping and notice the big REO AUCTION signs, or the huge display ads running constantly in the local paper now. Buyers do not need much more encouragement to procrastinate these days. This may be the reason why pending sales in April so far is about 22% lower than April 2006.
Lag time – batching properties for auction typically takes 30-60 days, if not more. The auction promotional period is another 30 days. If the marketing conditions are rapidly deteriorating, you are giving up valuable months.
Ineffective – the best buyers are always the owner-users. They usually come from local brokers who have already narrowed down the submarkets for the buyers. Mass marketing often attracts the wrong buyers. In residential real estate sales, you do not need a large number of lookie loos; you only need a few buyers most suited for a particular home.
Limited use – auctions are most effective for non-repeating final liquidations. Just like new home sales, auctions can create “death spirals” where incentives beget more incentives beget price discounts beget more price discounts.
I called a couple of local REO brokers to get a feel of what the business is like these days. Here is what I heard:
Hearsay #1 – early or first payment defaults are truly intentional. Borrowers maxed out on a cash-out refinance and then simply wait for the foreclosure process to finally end when the Marshall removed them from the premises, often a year or more from when they first stop payment. This behavior is not only irresponsible but it surprised me that so many are knowledgeable enough to choose this way out.
Hearsay #2 – all REO brokers have more business than they can handle, even as more brokers are reclassifying themselves as REO “specialists”. Based on the numbers that I was told, REO volume has indeed surpassed the worst of times in SD since the early 1980s.
Hearsay #3 – asset management companies are more prevalent than previous cycles. In other words, lenders/servicers outsource REO disposition to a third party who then contracts with individual brokers where the REOs may be located. These asset management companies are swamped nationwide, hence the auctions in hope of disposing properties in volume.
Hearsay #4 – REOs had been clogged in the pipeline because lenders were slow in reacting to the rapidly declining market. They are now at a point of realization and are adjusting prices accordingly. If this auction proves to be disappointing, the lenders are likely to be far more aggressive.
Modern day moral hazard – ironically, a subprime borrower could be a huge benefactor, even if undeserving. For example, an EMPLOYED subprime 55% DTI “homeowner” defaults and stops paying the mortgages, which is really just rent anyway, receives an immediate debt relief. Then it is free housing until the eventual lock out before moving into something similar as a renter for probably half the expense. DTI drops to below 40%. They may also receive a generous work-out offer, a package that is not available to the best of the prime borrowers and get to keep the home.