Bottom Testing in Mortgage Land « naked capitalism

Monday, May 12, 2008

Bottom Testing in Mortgage Land

The Financial Times has a long analysis, “Value to Unlock,” on how various financial players are starting to pick and choose among distressed mortgage assets. This may prove to be premature (recall how Wall Street firms eagerly bought subprime brokers through January 2007) and the housing market itself appears unlikely to hit its floor before 2010, but even at this stage, there may be pockets that are so cheap that the downside is (or appears) modest.

Note this seems to be a theme du jour: Calculated Risk also saw signs of buying in some distressed markets, and too is cautious about reaching conclusions.

The set up is that the Mortgage Bankers Association, a sober and sparsely attended affair this year, is galvanized by the report that a BlackRock fund will buy $15 billion of UBS’s subprime debt for 75% of face (UBS is a minority investor in the fund; query whether there was more than meets the eye here).

The piece comments on various players buying servicers allegedly to gain expertise. Maybe I am talking to the wrong people, and Tanta et al will correct me, but my sources tell me servicers are factories, and don’t have skills that are relevant to evaluating mortgage pools. Remember, the credit decision was made long before the securitized mortgage got in the hands of the servicer (and in the recent environment, calling them “credit decisions” is generous. More accurate might be “handing out cash to anyone who had a pulse and could fill out a form”). I’m a little perplexed that this factoid, while narrowly accurate, keeps being bandied about uncritically.

From the Financial Times:

Policymakers, bankers and investors all want more buyers like BlackRock to emerge for mortgage securities and other risky assets, to provide a tipping point that ends the credit crisis. Yet thus far there has been only patchy evidence that this healing wave of purchases is under way.

Certainly, the market for corporate debt has shown some positive signs. The $23bn buy-out of Wrigley by Mars, agreed two weeks ago, involved more than $10bn of debt – although less than $6bn of that debt came from investors, with the rest provided by Warren Buffet’s Berkshire Hathaway.

The debt capital markets, which last year made possible deals that were twice that size, still have a long way to go before they recover fully…..Banks are so desperate to rid their balance sheets of these loans that they are offering their best clients sizeable discounts and lots of leverage to sweeten the sales….

Investor fear of buying distressed assets too soon and catching the “falling knife” is even more intense in the mortgage market, where premature buyers have been badly burnt by plummeting asset prices…

Nevertheless, BlackRock’s deal with UBS represents one of the more significant examples of a small but growing number of contrarian bets on distressed mortgage assets by opportunistic buyers. Goldman Sachs, TPG and other investment banks, private equity firms and hedge funds have also started looking to buy portfolios of mortgages – in some cases reversing bearish bets made last year.

Indeed, in the past 10 months more than 80 funds have begun raising money to buy bad mortgage debt on the cheap, including Marathon Asset Management, GSC Group, Pimco and Fortress Investment Group. Goldman is trying to deploy around $4.5bn to invest in mortgage assets…

Placing values on distressed mortgage assets remain an enormous problem for both buyers and sellers, in part because it is hard to predict the full extent of the continuing slide in US home prices and the accompanying level of mortgage defaults and foreclosures.

Part of the difficulty is that faith has been lost in the measures of probable losses on which lenders used to rely, such as credit ratings and historical data. For pools of subprime mortgages, guides such as loan-to-value ratios, used to compare the size of a loan against the value of the property on which it is secured, have proved unreliable.

Mark Fleming, director of economics at First American CoreLogic, a research provider, says that while reported loan-to-value ratios for many subprime mortgage pools had been around 100 per cent, the existence of unreported “piggyback” loans meant that in some instances the ratio could be as high as 160 per cent.

“The problem is that while market-based pricing is not necessarily commensurate with the true risk, it’s still hard to measure the mismatch between pricing models, rational pricing opinions and prices driven by fear,” says Mr Fleming.

The valuation problem is worse for more complex instruments, for which there are still no buyers, particularly if these fall under fair-value accounting rules that require securities to be “marked to market” – priced on the books at no more than what is achievable. Susanna Kondraki, vice-president at Risk Span, an advisory firm, says one client spent $250,000 on valuation services for a $1bn portfolio of collateralised debt obligations backed by mortgages, only to discover that the portfolio had to be valued at zero.

James Fratangelo, head of whole loans sales and acquisition at Bayview Financial, says snags like this are why many parts of the mortgage market are yet to establish clearing prices. “There is plenty of liquidity for distressed assets but there is still a huge gap between where buyers want to buy and where sellers want to sell,” he adds – with the gulf between bids and offers remaining as wide as 20 cents on the dollar for many assets.

Robert Gaither, head of the secondary marketing group for mortgage securities at Bank of America, illustrated this problem at last week’s conference. He described receiving bids for a portfolio of so-called “Alt-A” mortgages, between prime and subprime, that the bank had marked down to 91 cents on the dollar. After a series of bids from prospective buyers at 50 cents, he finally received one at 86.5 cents. Yet the bank’s pricing model said the mortgages should be priced in the mid-90s.

The bank decided to hold on to the portfolio, even though it was forced by mark-to-market accounting rules to write down the mortgages to match that 86.5 cent bid.

Many European banks are also refusing to sell at prices that they consider to be artificially depressed….

She says European banks’ belief that such asset prices have fallen too far has been bolstered by a recent Bank of England report suggesting that triple-A tranches in particular were mispriced……

Still, there are signs – including UBS’s sale of loans to BlackRock – that some higher quality assets are beginning to move…Traders say the scale of buying generally remains small, however. In the European markets, buyers are placing orders of just €20m (£16m, $31m), far below the €500m orders that were normal before the crisis broke.

Many funds say they remain constrained in the volume of deals they can do by the sheer difficulty of raising finance. Others fear this means there is too little capacity to absorb the volume of distressed assets in the market.

“What worries some people is that you have seen a few people fill up but it’s not clear whether there will be more buyers after that – it could just be one or two groups that are ready to buy,” says one London-based hedge fund official. “The market is so thin and prices are so volatile that if they stop buying, we could go back down again.”….

One of the biggest problems facing prospective buyers of distressed mortgage assets is that US house prices continue to fall and the flood of late mortgage payments and foreclosures shows few signs of abating. Unless the rising tide of losses in the housing market can be stemmed, establishing a floor under asset values may be difficult, writes Saskia Scholtes.

Mark Kiesel, a portfolio manager at Pimco, the big US bond investor, says: “We may need housing prices to bottom for this entire process to trough and for most markets to rebound.”

19 Comments:

Anonymous says:

May 12, 2008 at 3:38 am

Where or where is Joseph Welch to ask

You’ve done enough. Have you no sense of decency, sir, at long last? Have you left no sense of decency?

BlackRock advises the FED on the $30 billion (excuse me, $29 billion) ‘assets’ in the JPM/BSC ‘non-bailout’

And BlackRock accquires equity interest in very similar assets.

What do they know that others don’t?

Ginger Yellow says:

May 12, 2008 at 4:47 am

All the usual caveats about Gillian Tett articles apply (the paragraph on the non-existent “Carlyle Credit”, for instance, is riddled with basic errors). That said, the point is that you don’t just buy a distressed asset, sit on it, and hope everyone else is wrong. You buy it and then work damn hard to make sure everyone else is wrong. You need servicing expertise to turn the asset around. Servicers may also help in the valuation/acquisition stage by identifying pools of mortgages that have been serviced really badly, and hence could improve considerably in performance with some care and resources thrown at the problem.

Yves Smith says:

May 12, 2008 at 5:07 am

Ginger Yellow,

Agreed re Tett, She has gone downhill since she was promoted (roughly a year ago, I think).

Re servicers: what I hear from people I know who work in the subprime arena, but from a different vantage (mortgage counsellors, who recall had to be brought in as part of the Hope Now program) and not for profits that lend to the same type of borrower and whose losses are the same as on prime mortgages, is that the servicers know bupkis about their borrowers. Not only are the loan files generally useless (inaccurate and incomplete) but even when they have decent info, the servicers lack skills to evaluate them (I have been told that emphatically by more than one person).

Serivcers are factories. Staff roles are highly specialized. They collect payments, keep records, remit payments to investors. They were never intended to do loss mit in a meaningful way; remember, borrowers could refinance their way out of trouble.

And servicers are also mistrusted by the borrowers, so even if they somehow acquired the skills to do mods, they’d have trouble getting the borrower’s trust to get sufficient disclosure.

The other ugly fact is servicers are hemorrhaging cash (I heard of one that was losing a billion a month). The servicing agreement require them on default to continue paying interest for 90 days, in some cases principal too. In all cases, they have to pay real estate taxes and insurance, and that obligation goes beyond 90 days. Yes, they can attempt to get it back (plus fees) from the deadbeat, but good luck. Their servicing agreements provided a cushion for an expected level of defaults, but with defaults so high, they’ve blown through them.

It would thus be in a seller’s interest to persuade bottom fishers that the servicers had insight to get them to take this liability off their hands.

With that fact set, I am not certain what you mean by “a bad job of servicing.”

bobo7874 says:

May 12, 2008 at 5:56 am

anon 4:38am,

There are significant differences between the assets Blackrock will manage for the Fed/JPM, and bank loans (even levered ones with light covenants). The latter are significantly easier to value and totally different investments; hell, even pools of mortgages are significantly easier to value than asset-backed securities. You concerns are unfounded.

As an aside, in part asset-backed securities are hard to value because of Fin 46 accounting rules that force the issuer to be a passive dummy entity in order for the sponsor to avoid consolidating it. The banks lobbying for the Fin 46 rules that allowed them to deconsolidate issuers of asset-backed securities, as well as their lobbying related to SEC Rule 3a-9, have helped banks make oligopoly profits in securitization. As soon as you force an issuer to be a dummy entity, it vastly increases the computational difficulties in matching the underlying pool of instruments held by the issuer with the instruments issued by the issuer. The banks had an advantage in doing the math over others.

The banks also had an advantage over others due to Par 9(c) of Fin 46. FASB and the SEC blessed the banks’ position that they didn’t have to consolidate an asset-backed commercial paper entity they’re sponsoring (at least on formation) because the bank’s investment in the entity didn’t exceed the “estimate” of the entity’s “expected” losses. I’ve heard this referred to as the “expected loss tranche loophole”. I thought the rule is in Paragraph 9(c) of FIN 46. I thought FASB and the SEC ended up approving of this practice.

One has to respect the banks lobbying. It is highly effective.

bobo7874 says:

May 12, 2008 at 6:04 am

There is a typo in my prior post, I meant SEC Rule 2a-7, not Rule 3a-9. Apologies.

Tanta says:

May 12, 2008 at 6:40 am

Yves, I wouldn’t say that no servicer has the ability to do pool/portfolio valuations. The thing is, exactly, if you are a servicer, you buy servicing rights to loan pools. You kind of, you know, have to. If you are not also an originator, you need material to work with.

So these outfits that are not originators buy servicers, then they will indeed buy loan servicing, whether they also managed to acquire the expertise to evaluate purchased servicing rights adequately or not. Whether now is a good time to buy servicing or not. You buy a servicer, you are going to have to buy servicing. The time to evaluate the quality of the servicing rights on offer in the marketplace is before you bought a servicing operation to feed.

The street got into buying originators a while back because, well, that’s actually the best way to provide yourself with an on-going stream of servicing rights you know something about and can control the quality of. Why didn’t it work? Because you have to know something and control quality to be a successful originator. Some of them seemed to have skipped that part. If you don’t originate your servicing rights, you buy what someone else originated, and you put a ton of due diligence into the acquisitions or else . . . you get what we have today.

What you say about cruddy servicers is true, and is driven by the need to squeeze profits out of a low-margin business. They have these models that tell them how this can be accomplished by paring expert staff down to the bone and not caring about the file contents and skimping on loss mit and stuff.

Those would be the valuation models these folks would be buying.

Ginger Yellow is right that the only way you can make any money buying distressed mortgage assets is to own a “combat servicer” that can indeed improve the value of those assets by remedial servicing work. But the big problem there is, of course, what you pay for your “combat servicer” as well as what you pay for the assets.

Ginger Yellow says:

May 12, 2008 at 6:43 am

Bobo, I’m not sure what your point is about FIN 46.

a) How does para 9(c) give banks an advantage over other entities? The rule applies to all institutions. It’s just that banks tend to be the ones providing liquidity to ABCP, because providing liquidity is what banks do (among other things), while other entities tend to take the first loss these days because of the rule. Usually nobody ends up consolidating an ABCP vehicle, however, because the expected loss is shared so that no one entity has exposure to a majority of it.

b) I’m not sure why you think having a dummy company as an issuer in itself makes valuation difficult. The difficulty comes in modelling the cashflows of the underlying assets (equally difficult whether a naked pool or a securitised one) and in calculating how the structural features of the securitisation (eg the waterfall, any triggers), affect the cashflows to a given tranche under different scenarios. This is true regardless of whether a dummy company is used. The point of the dummy company requirement was to limit the non-consolidation and derecognition benefits to true securitisations rather than Enron style frauds.

c) FASB is in the process of revising FIN 46 and FAS 140 to remove QSPE status for securitisation vehicles. This will almost certainly mean consolidation for sponsors – the most favoured option at the moment is a linked presentation system similar to what the UK had before it adopted IFRS.

Yves: “With that fact set, I am not certain what you mean by “a bad job of servicing.”"

Well, how about not contacting borrowers after the first delinquencies, poor follow-up of delinquency, no attempt at loss mit, understaffing, etc. Calculated Risk has highlighted a lot of stories where servicers seem to have failed at the most basic tasks assigned to them, possibly because of the pressures you describe. If these distressed buyers have any sense, they’ll acquire a servicer with experience and skill in special servicing, not a general purpose servicer.

bobo7874 says:

May 12, 2008 at 9:23 am

Ginger Y,

There are big competitive advantages the banks have as sponsors of conduits.

The Fin 140 rules preventing QSPEs from purchasing and selling assets have but requiring them to
show that their assets are continuously matched to the aggregate obligations under beneficial interests increases the complexities involved in sponsoring conduits, giving big financial institutions an edge by increasing fixed costs and helping those that sponsor multiple conduits.

Another edge the banks have is that within these tight parameters, the party given control of the conduit needs to be sufficiently credit-worthy (heh) to trust it with discretion to harm the value of residual interests it does not consolidate. This gives an edge to financial institutions with gilt edged credit ratings, like some special sub with a AAA rating.

Rule 2a-7 helps the banks because the 10% obligor rule increased structural costs by forcing money market funds to buy more pools of loans, rather than a single large diversified pool. In the mid-1990s, for example, REITs sponsored large diversified pools, but no more. The increased structural costs help banks because they sponsor many pools and can spread the costs among them.

Par 9(c) helps the banks too by allowing them (or as you point out others) to avoid consolidation. The “expected loss” calculations seem flawed in a number of respects, given the extent to which the generalize from past experience without hard thinking about the future.

Ginger Yellow says:

May 12, 2008 at 9:56 am

But someone’s got to provide the conduits with liquidity, or they’d fall apart like the SIV market has. And who’s going to do it if not a bank? I’m still not seeing what your big picture point is here. Conduits are basically off balance sheet banks. Of course banks are going to be the main sponsors – surely that’s good and proper. I’d agree that banks gain a competitive benefit from being able to park assets in a conduit, but I don’t see them as being particularly advantaged among likely conduit sponsors. At least not for the reasons you give – banks tend to be more highly rated than non-banks, which is obviously an advantage.

“The Fin 140 rules preventing QSPEs from purchasing and selling assets have but requiring them to
show that their assets are continuously matched to the aggregate obligations under beneficial interests increases the complexities involved in sponsoring conduits, giving big financial institutions an edge by increasing fixed costs and helping those that sponsor multiple conduits.”

But what’s the alternative? The whole point of securitisation vehicles is that they’re bankruptcy remote. If you introduce the possibility of insolvency at the SPE level, you take away the basis for the entire securitisation market.

bobo7874 says:

May 12, 2008 at 10:51 am

Ginger Y,

Historically, there’ve been at least 4 types of conduit structures: diversified-support conduits, securities arbitrage conduits, single-support conduits, and unitary support conduits. FASB and the SEC through 2a-7 and other rules gave single-support conduits an edge over the structures previously used by REITs and other sponsors. Using the other structures wouldn’t kill securitization, just increase competition for banks, and reduce their profits.

Ginger Yellow says:

May 12, 2008 at 11:20 am

Sure, but allowing securitisation vehicles to not be dummy companies would kill it. Also, I’m not familiar with your terminology. ABCP conduits are conventionally broken down into securities arbitrage, single-seller, multi-seller and SIVs, based mainly on the underlying assets (SIVs being the excetpion). Your terminology seems to apply to the liquidity support, however. Is that right?

I’d also add that a) Basel 2 greatly reduces the capital advantages of conduits for banks, and so will level the playing field in itself when introduced in the US, and b) Reits and other non-bank sponsors will still be disadvantaged in that investors are paying a lot more attention to the liquidity providers than they used to. A strong bank sponsor will be able to fund its conduit far more cheaply than a non-bank sponsor or a weak bank until we get into another mad bull run.

bobo7874 says:

May 12, 2008 at 8:42 pm

This is my terminology:

Diversified support conduit = liquidity & hedging support from multiple third parties, ie, deficiency obligation for third parties.

Securities arbitrage conduit = credit & liquidity support by a seller; hedging support by a seller, ie, deficiency obligation for a seller

Single support conduit = credit, liquidity & hedging support by the sponsor, ie, deficiency obligation for sponsor

Unitary support conduit = credit, liquidity & heding support by the originator or related persons, ie, deficiency obligation for originator

2a-7 and other rules favor single support conduits, which is the structure where banks have an advantage. The preferred treatment for QSPEs under Statement 140 over VIEs under Fin 46 also helps banks by reducing discretion and thereby increasing complexity and need for a highly creditworthy sponsor.

Ginger Yellow says:

May 13, 2008 at 4:01 am

Thanks for the clarification, although if anything it makes your point even more obscure to me. The vast majority of conduits these days would seem to fall into the diversified support category – liquidity and hedging support from one or more banks, credit enhancement from third parties (programme wide) and the seller (asset specific). Which suggests the competitive advantage for “single support” conduits isn’t that strong. Indeed, given that a single support structure gives all exposure to expected loss to one party, I’d argue its least favoured under current rules.

bobo7874 says:

May 13, 2008 at 5:53 am

Thanks, you’ve persuaded me that the banks’ competitive advantage due to FASB & SEC works differently than I’d thought. But I am pretty confident it exists and that is why select financial institutions have dominated since the 10% obligor rule and other changes. I need to think about this some more.

Ginger Yellow says:

May 13, 2008 at 8:01 am

I’m struggling to envisage a world in which financial institutions didn’t dominate. Conduits borrow short and lend long to finance assets, which is exacly what banks do. Now often those assets are trade receivables, but it makes far more sense for most companies to finance your assets in a bank sponsored conduit (where investors can be more confident the sponsor will be around to provide liquidity) than to set up your own. You’re still going to need liquidity support from a highly rated firm or you won’t get a good rating on your CP.

bobo7874 says:

May 13, 2008 at 9:42 am

Of course financial institutions dominate. The question is whether “select” financial institutions dominate at the expense of smaller players like REITs, and thereby increase financing costs for the real economy. When I said “select financial institutions” in my prior post, I was thinking money center banks and prime brokers. And I do think the 10% obligor rule hurt REITs and other smaller financial entities.

Ginger Yellow says:

May 13, 2008 at 11:49 am

But why should Reits compete equally here? An ABCP conduit seems like a spectacularly bad funding structure for an institution investing in lumpy, relatively illiquid long term assets like real estate. It proved bad enough for the sec arb conduits investing in smaller tranches of things like CMBS. Surely debt finance for real estate should be long term, not commercial paper.

Ginger Yellow says:

May 13, 2008 at 11:51 am

Moreover, I think any impact of the rules in either direction is far outweighed by investor sentiment. In this current market, and for the foreseeable future, “select financial institutions” are going to have a massive advantage finding investors for paper they sponsor over other entities, regardless of accounting treatment.

bobo7874 says:

May 13, 2008 at 4:29 pm

I don’t necessarily care about REITs in particular, what I do care about, is FASB and SEC rules providing financial/securities rules that help too-big to-fail-institutions over other financial entities. Check out the credit spreads of AA over high yield post-1990 when major changes were made to 3a-7, 2a-7, grandfathering of old instruments against the 1998 modification to 2a-7. When these institutions received financial/securities rules putatively favorable to them, credit spreads increased. Conversely, when their putatively favorable financial/securities rules were weakened credit spreads shrank. That suggests to me that these putative advantages help them, and increase financing costs to the real economy by reducing competition by other institutions.

Subscribe to: Post Comments