How I long for the days when MBS research was about dry, specialized stuff like prepayments and relative value, projecting cash flows, building data bases and models of prepayment and yield curve processes. Dull matters you would be careful not to mention at a dinner party or drinking with pals in a bar. Tedious topics that could put Main Streeters to sleep. Conversation stoppers.
That changed in 2008. Mortgages and houses and mortgage-backed securities have transmogrified into filthy politics. Nowadays, opinionating about mortgage markets is a free-for-all with no barriers to entry, in which unsupported theories and poorly reasoned malarkey are spewed by politicians, no-experience academics, economists with no mortgage experience, the dying print media, the new electronic media, the blogs and the social nitworks.
The idea is not to monetize good information, but to get attention, attract site traffic, raise great winds of popular response. To stir the pot.
Problem is, that little coterie of data-respecting nerdalysts of which I was once a member now must spend an inordinate portion of the work day dissecting government half-measures (Hope Now, HAMP, HARP and so forth ad infinitum) and weighing rumors of more on the horizon.
With increasing frequency, nervous institutional investors call on the professional analysts for a reality check on wildfire speculation, canards and outright lies about what the government intends to do to fix the housing-mortgage mess and, unintentionally, sabotage investment performance.
The worst patch yet, I think, has been seen in the last couple of weeks, thanks to deliberate rumormongering about a backdoor stimulus plan
, in which the government compels the housing agencies to forgive principal on millions of underwater mortgages and refinance them into the lowest published mortgage rates on record. Rumor mongering intended to further confuse public discourse going into the elections.
Plausible Ain't the Same as True
There are lots of reasons for believing such a plan would exist. Millions of American households are underwater on their mortgages. Delinquency, foreclosure, and vacancy rates are at record levels. Programs launched by two Administrations have made only a modest dent. Headline mortgage rates are blindingly low, but refinancing activity is very subdued because borrowers don't meet Fannie and Freddie's (the GSEs') credit requirements or can't afford the risk-adjusted rate. Obstacles include mortgage debt bigger than the house appraisal and sagging household incomes and credit scores.
The Home Affordable Refinancing Plan (HARP) -- intended to streamline GSE mortgage refinancing for underwater borrowers -- has not been a success. And faced with losses to the insurance fund, even the Federal Housing Administration
(FHA) has had to tighten its streamlined mortgage criteria. Some homeowners have decided the rational strategy -- even if they can afford their current mortgage -- is to default, to put the mortgage (and the house!) back to the lender. The economy is treading water, generating much discussion of double dips and deflation and turning Japanese, but it is widely assumed additional stimulus spending will not carry the Congress. And, with broader financial reform now enacted, the Administration is moving to address the future of the housing GSEs.
And the kicker is that the mortgage industry -- the origination/servicing apparatus that feeds and waters the government-enabled housing finance system -- has been able to dramatically consolidate, gift of the crisis, into the hands of the too-big-to-fail banks. The scrappy competitors are gone and the survivors are making more money than ever on new originations (see for instance "Originators Lend Less But Profit More on Mortgages", Kate Berry, American Banker, July 26, 2010).
Professional Investors on Edge
As much as pols, pundits and bloggers expect these conditions to provoke a BIG government response, the MBS market fears it. Things have been going almost too well for pass-through investors.
For example, in a July 16 report, JP Morgan
mortgage analysts Matthew Jozoff and Brian Ye reviewed government policy options in the wake of HARP's dismal performance. They led by saying: "With mortgage dollar prices soaring to record highs practically each week, investors' faith in 'prepay nirvana' is being tested to an unprecedented degree. The looming fear is a government sponsored refinance wave, eliminating the many hurdles that have stood before prospective refinancers and pushing speeds back to 2003 levels."
(Let me explain: A refinancing wave hurts investors two ways. Prepayments erode expected returns on bonds priced above par; even before prepayments materialize, market prices fall. Prepayments return as supply, for which the audience, frankly, has shrunk. Investors lose again as supply cheapens the sector.)
The JPM analysts worked through a few stratagems that might spur a 2003-style refi wave, each problematic for good practical reasons, which they explained in some detail. They concluded any government policy to stimulate a mass refi wave would be difficult to implement (if it weren't it would have been tried before). Furthermore, benefits for borrowers in the short run would be offset by higher long term borrowing costs.
Chicken Little Time
The JPM report, because it did not support the concept of a government sponsored refinance wave, went unnoticed outside the MBS market. Inside the market pass-throughs continued to ride high. That is, the end of July, when a Morgan Stanley
economist put an M-80 in fixed income investors' mailboxes.
Calling it a "Slam Dunk Stimulus" (July 27), David Greenlaw proposed wholesale refinancing of all government-related mortgages with above-market mortgage rates, without regard to current value of the underlying property or borrowers creditworthiness.
Let me emphasize that Greenlaw is not a mortgage market analyst. He has the same grasp of the salient details, the organic mechanisms that funnel capital markets cash into home loans as, for example, your average real estate agent. Close, but no cigar.
The way I see it - and I am standing on more than twenty years of MBS/ABS research experience -- the only reason this report received the attention that it did is because it was sent, not to experienced MBS market participants, but to a broad general audience.
Because this is how it works, regardless of the subject, whether it's health care legislation or the role of the GSEs: any exaggerated assertion will go viral if made to people with no specific information on the subject.
(And what would something labeled "slam dunk stimulus" be, if not a flashy exaggeration?)
Let Fannie, Freddie (and FHA) Distribute Cake
In the piece, Greenlaw recommended that the government streamline the refinancing process to allow all borrowers with high LTV mortgages behind Fannie, Freddie and Ginnie MBS to refinance without analysis of the borrower's current FICO or income verification. Says Greenlaw, "There is no need for a case-by-base analysis of a borrower's credit quality when the principal value of the mortgage is already backed by the government."
In effect the government would "merely" recognize the existing guarantee on a huge part of the mortgage market. He makes it sound so simple!
Here's Greenlaw's math, borrowing some round numbers from Morgan Stanley mortgage strategist Janaki Rao and housing analyst Oliver Chang: If an estimated 18.5 million mortgages - half of the government-guaranteed market - refinanced to a current rate (mortgage rates are at new all time historic lows), that would generate annual savings of $46 billion per year.
Plowing those savings back into consumer spending would be one heck of a fiscal stimulus.
Greenlaw concedes that, while the household sector would benefit, holders of agency MBS would be losers. Prepayment speeds would rise dramatically and MBS spreads would adjust accordingly. This could offset some of the gain, so maybe the Fed should step in and again provide the backstop bid to the market.
If an economist colleague had ever approached me on the trading floor with this idea I would have said, "WHAT ARE YOU SMOKING?" Bear with me a bit longer, and I will lay out good reasons, from good MBS specialists, why Greenlaw's idea, as presented, is not feasible.
MBS Market Rattled
Greenlaw's report bounced around fixed income markets on July 28th. MarketNews International, for instance, noted the report in its news bullet service to subscribers midday on the 28th and end-of-day market commentary from various dealers indicated mortgages, especially higher coupons, had underperformed benchmarks.
Bloomberg didn't catch wind of the Morgan Stanley report until the next day, when Jody Shenn at Bloomberg mentioned it inside another story on hedge fund manager Deepak Narula's concerns the US could change the GSE rules and grant a "one-time amnesty"
In the meantime, I can just imagine what was transpiring in the pass-through market, having been through these fire drills myself. Investors pass the report on to any MBS traders and analysts they talk to regularly. What, they want to know, is the likelihood something like this could happen?
Within 72 hours bona fide MBS analysts in at least half a dozen brokerages were driven to publish their take on a "government-induced refinancing wave".
Said Barclays' Ajay Rajadhyaksha (italics are theirs):
• There has been no mention from Washington of a new refinancing program "even as MBS investors have worked themselves into a frenzy";
• Banks would benefit as well; the actual transaction would require the GSEs to waive its option to put bad loans back to the lenders, opening "policymakers up to charges that big banks that made bad loans are being bailed out, again";
• Actual refinancing rates would be pushed up, lowering the potential savings to $5 or $6 billion a year (not Greenlaw's $46 billion);
• Lacking standard credit underwriting, the new loans probably can't be pooled in TBA; uncertainty about how far behind TBA the new MBS would trade could reduce participation by lenders;
• A solution to these problem would be for the Fed to commit to buy MBS backed by the refi loans, while selling a matched amount of its existing MBS holdings into the market. Barclays MBS analysts also believe the refi-backed securities will have less extension risk should rates rise.
refinancing is unlikely", according to Bank of America/Merill Lynch
analysts Chris Flanagan, Vipul Jain, Vivek Sriram and Virkram Rao. They see it as equivalent to partial default in theoretical terms, but practically speaking they aren't sure whether it would be structured as a modification or a refinancing. Approaching it as a modification raises legal problems with the trust documents: "as they stand now they do not allow buyouts of loans that are current". On the other hand, the refinancing strategy has already been tried by HARP. Futhermore, lenders are already lending close to their origination capacity. Their normal response to an uptick in volume is to raise rates, reducing the benefit to borrowers. Plus, additional supply puts pressure on mortgage spreads. As these pressures raise rates, the number of borrowers that can benefit from any government program declines.
The BofA/ML analysts dwell as well on put-back risk. Lenders are reluctant to refinance high credit risk borrowers serviced by other lenders because they bear put-back risk and because, if the loan goes bad down the road, servicing delinquent loans is very costly. Moreover, modification programs have made those servicing costs uncertain. Pondering these issues, BofA/ML analysts think lenders will only refinance loans from their own servicing books, fracturing the effectiveness of any program coming from Washington.
Even assuming the mechanics could be worked out, BofA/ML analysts argue the program cannot be costless to taxpayers. A huge increase in origination capacity would be required, requiring new financial incentives from the government to induce lenders to participate.
analyst Mahesh Swaminathan said, "forget a 'slam dunk', a government-induced refi spike is not event a 'lay-up'". One thing Swaminathan considers other analysts do not, is the difficulty of designing simple enough criteria to facilitate mass refinancings, but do not over-subsidize whole subsets of borrowers. Borrowers' cash constraints have to be considered as well - the government might have to roll origination fees, points and costs into the loan (at a future cost of higher losses on loans that eventually default). Suppose the government succeeds? Then who buys as much as $750 billion in new lower coupons? The Fed. "This faces huge opposition from policy makers and would complicate an exit strategy." And it would remove even more tradable float from the market.
Nomura analysts Ankur Mehta, Dhivya Krishna and Ohmsatya Ravi see options. The government could eliminate loan level pricing adjustments (LLPAs; these are risk-based pricing adjustments that the GSEs first instituted in late 2008 and have been refining since). Other options include eliminating income verification and employment documentation, loan-to-value and debt-to-income requirements. However, current investors "would suffer meaningful losses":
• The Fed and Treasury own close to $1.3 trillion agency MBS purchases at an average dollar price of $101 to $103. They are not marked to market, and any losses realized are likely to be less than earnings to-date on carry;
• US banks and savings institutions own $1.15 trillion agency MBS. They would be forced to mark down those holdings by upwards of $58 billion - a direct hit to Tier 1 capital ratios. However the impact on net income would be much lower - the difference between amortized cost and an estimated average dollar price of $101 after the changes;
• Overseas holders would see the market value of about $700bn agency MBS decline by about $35bn;
• GSEs own about $650 of their own MBS and would lose about $32.5 billion. On the offset, Nomura says credit losses might decline - by one estimate about $21-$22 billion;
• Insurance companies and state/local governments own $400 billion and $250 billion, stand to see mark downs of $20 billion and $12.5 billion, respectively;
• Another $800 billion in the hands of mutual funds, REITs, hedge funds and households could see mark-to-market losses of $40billion;
• Pressure on mortgage spreads and higher Treasury yields could push the primary mortgage rate up to 5 - 5.25%, leaving refinancing attractive to borrowers backing 5.5s and higher (that's about 45% of the 30-year universe).
Without a double-dip, Nomura sees less than 10% probability of any Govt/GSE actions. If the double-dip materializes, the probablity might rise to 30%.
analysts Brett Rose and Inger M. Daniels argue the unintended costs of a "government-induced refi poof" would include a $30 billion premium loss in the GSEs retained portfolios, a $5 to $10 billion increase in Treasury borrowing costs with higher resultant yields, 100 basis point higher mortgage rates, possibly pushing home prices down 10%. High coupon, shorter duration securities would be replaced by low coupon, long duration securities, sharply increasing the hedging requirements of the MBS universe. They think a 25-50 basis point rise is a reasonable estimate.
Curiously, such a program would not do much to improve foreclosure rates, say the Citi analysts. The explanation is quite simple - perhaps 90% of the loans that might be included are not delinquent.
JPM analysts revisited the topic in their most recent weeklies. Writing July 30, they too worried about the "nuts and bolts" of implementing such a plan and the hidden costs of any forced "refi tidal wave". They also weighed more "conventional" options. The most obvious is to eliminate the GSEs Loan Level Pricing Adjustments (LLPA) for refinancings. But they estimate eliminating them would generate only a modest refi response in the most credit impaired loans. Raising the LTV limit on HARP refinancings from 125 to 150% is another possibility. However, only about 4% of conventional borrowers have current LTVs above 125%. Significant impact could be achieved waiving reps and warrants, which would eliminate put-back risk, but JPM assigns low odds that the GSEs and their regulator, FHFA, would go for that option. The government could direct originators to solicit borrowers to refinance and reimburse the origination costs. Finally, the 2nd lien hurdle could be removed although "there are significant legal barriers to taking this course". (Actually, I think Congress has already addressed this "hurdle" with little effect, but that's a topic for another day.)
In their August 6 weekly they attacked the idea of principal forgiveness head on. Estimating that 20% of the conventional universe is underwater an average 10%, that means $1 trillion of the $5 trillion agency MBS universe has $100 billion negative equity. That is a very big check for the government to write in a deficit-conscious environment.
MS Means Missing Something
Housing economist and MBS market veteran Tom Lawler also responded to Greenlaw's proposal as "MS - Missing Something!!!!" Tom's thoughts are usually subscriber-only, but he shared this commentary with Calculated Risk
Greenlaw appears to have missed the fact that the government has already used the argument that the guarantee is already in place to significantly expand the GSE refinancing window with HARP. The GSE needed this argument to liberally interpret the requirement, in the charters granted the GSEs by Congress, that loans with LTVs above 80% have third-party credit support in the form of private mortgage insurance, lender participation, etc. FHFA used the argument to rule that loans originally having LTVs of 80% or less could be refinanced at LTV's above 80% without new mortgage insurance. The Treasury claimed it again with "the well-intentioned but poorly executed" HARP program, which first allowed CLTVs up to 105%, now 125%.
In fact, Lawler thinks it "astounding!" that Greenlaw doesn't mention HARP. Nor does he mention the FHA's streamline refinance program, which Lawler called "an inconceivable miss on their part!!!!" Another of Greenlaw's ideas that has already been tried.
Lawler also illustrates, in very explicit, real-time detail how little Greenlaw understands "the refinance process, transaction costs and mortgage rates". These are the cash requirements Swaminathan points up. In Lawler's example, the 4.5% current rate Greenlaw quotes is very close to the current Freddie Mac Primary Mortgage Market Survey average 30-year rate. The average borrower at that rate paid an average 0.7 points. There are closing costs in addition. Working with information provided on one big lender's website, Lawler estimates that points and closing costs add over $6000 on a $180,000 mortgage in northern Virginia. If they are covered in the loan rate, Lawler estimates the real rate for a very good borrower is closer to 5.5% - a full point higher than the rate Greenlaw uses to project the $46 billion fiscal stimulus.
Now Its Political
All this rational discourse appeared to settle MBS markets. Participants and market-centric news media certainly did not take much notice on Tuesday August 3, when Richard Berner, Co-Head of Global Economics at Morgan Stanley, testified before the Senate Budget Committee. Among his recommendations was Greenlaw's "Slam Dunk Stimulus". Ho hum.
The peace did not last long. Like a pathogen harmless to domestic animals, but fatal to humans, measured discussion by better and worse informed, but still professional analysts jumped to the blogs. Where possibilities were reduced to actualities, blogheads and twitbrains substituted rumors asserting that the government was preparing a nuclear-option-style solution to bail out homeowners and goose the economy in time for the elections.
Maybe the tipoff was the August 4 MarketWatch story from Ronald D. Orol, "Could the government create a backdoor stimulus?" Orol asks upfront (unlike me, he does not bury the lede
), "An idea that could create a massive refinance wave -- and a back-door stimulus program to boost consumer spending -- is generating debate among investment bank analysts in New York and in Washington policy circles." He cites Greenlaw's paper, and a paper by equity analysts at Keefe, Bruyette & Wood
, Bose George and Jade J. Rahmani ("the costs are material and in the long run almost certainly outweigh the benefits" and HARP could be improved to "help underwater borrowers while limiting the collateral damage").
On August 5, James Pethokoukis, who posts under the heading "Political Risk" at Reuters.com, squawked "An August Surprise from Obama?
". According to Pethokoukis, "Rumors are running wild from Washington to Wall Street" that the Obama administration is about to order the GSEs to forgive a portion of the debt of millions of of Americans whose mortgages are underwater.
Pethokoukis goes on to heap political risk onto any bona fide policy making with malarkey like "if it happens [it] would be a stunning political and economic bombshell less than 100 days before a midterm election". And "Republican leaders believe this is going to happen". And finally, "Wall Street banks are alerting their clients privately to this possibility." And he quotes a Goldman Sachs
and a Mizuho
report, both written in economist-speak, remote from the realities of the mortgage and MBS markets.
This is great propaganda, but it's not constructive public discourse.
This drivel has spread all over the blogmire. Receiving Pethokoukis' piece from a number of readers the same day, Calculated Risk an oasis of thoughtful, fact-based commentary, pooh-poohed it
, "This nonsense is part of the silly season .... Alert Drudge and the tinfoil hat sites - they will run with this story. It is a political post ... I'm already sorry I mentioned it."
The Treasury duly denied the rumor, in a single line from Andrew Williams, a Treasury spokesperson, "The administration is not considering a change in policy in this area." Reuters ran the denial but it was like putting out fire with gasoline. A few hours later, like an electron accelerator, Pethokoukis re-upped his malarkey with a part 2.
"The Treasury Department has officially denied it is planning the mother of all mortgage bailouts. And I have no reason to doubt Team Geithner. But of course that assumes that the whole idea was not being cooked up by the White House political team (Rahm and Ax) and not the good folks at Treasury."
(Please read that carefully. Is that extra "not" a Freudian slip? The whole idea was cooked up NEITHER by the White house pols NOR by Team Geithner. Was it being dreamed up by venomous bloggers perhaps?)
But he goes on:
"During the financial reform debate, banking lobbyists continually complained that Geithner and Summers had been usurped by R&A in policymaking. And I have gotten zero pushback from the WH. Food for thought. More to come."
Huh? One - that was one hell of an ego scrum with G&S scrambling with R&A for the strings on the ponderous Dodd and Frank sheep dogs of the Congressional herd (or is it horde?). I think Pethokoukis has over-simplified the recent legislative miracle.
Two - I have been reminded by my publisher that Pethokoukis has been around the block with all kinds of publications, and does have WH sources, so I should give some credence to his assertion that the WH has not pushed back on his claims. But I am skeptical. One, why would you reward someone feeding bile to opponents of the Administration, on both the right and left (mostly right), with more than one sentence from "a Treasury spokesperson"? Two, how easy is it to convert the truth - WH ignored me - into "no pushback"? It's just words. He already works without facts, data, analysis.
Ah well. Time will tell. Pethokoukis says "More to come". And then he provides links to six "reactions" to his piece on other blogs. You can also get a feeling for where the infection has spread from looking at the comments to Parts 1 and 2.
All of them bereft of authentic information. My publisher takes one, at Atlantic.com, to task in his vent
. (The difference between me and Paul Jackson is he's a realist and believes the dumber the idea, the more likely pols will try to materialize it, and I'm a romantic who believes if you understood the facts of the situation, you would try surgery not bandages. Also, Paul doesn't bury the lede.)
You can find more - on blogs everywhere, from Time magazine to Rush Limbaugh. And of course, the Wall Street Journal. I've even found a link to a story that says "We're sorry this story isn't ready yet". Most of them wild with overreaction, righteous anger, malignant gratification. Use the obvious key words - "backdoor stimulus" or "stimulus surprise" or even (no imagination) "slam dunk stimulus". They're out there. Get used to it. Soon you'll be hearing the rumors, asserted as facts, by candidates in the upcoming elections, repeated endlessly on television and youtube videos.
NOTE: Linda Lowell writes a regular column, called Kitchen Sink, for HousingWire magazine.
Editor’s note: Linda Lowell is a 20-year-plus veteran of MBS and ABS research at a handful of Wall Street firms. She is currently principal of OffStreet Research LLC.