.comment-link {margin-left:.6em;}
Visit Freedom's Zone Donate To Project Valour

Monday, January 10, 2011

Recant Jimmy J, For You Are Doomed - Post 1

Jimmy left the following comments on a recent post:

There was no way to discover prices on illiquid MBSs that were constructed of many different mortgages. Only the mortgage sevicers knew whether individual mortgages were performing. No one KNEW what was in the MBSs. Therefore, all, botht the good and the bad, were bid down to 22 cents on the dolar. A price that was nowhere near the true value of the mortgages. Even if every mortgage in an MBS went into default, the underlying properties were still there and I know of no real estate market where prices have gone down 78%. Even with a total default the value should have been somewhere near 50 cents on the dollar. You cannot value an illiquid instrument like an MBS in an auction market unless there is some way of knowing for each MBS what the quality of the mortgages are, how many of the mortages are performing, and what geographical markets the mortgages are in. As far as I know that information is not available and there is no mechanism planned to make it available. That is why the FASB and SEC are allowing financial companies to carry the MBSs at a value arrived at by the percentage of income derived from performing mortgages. Not perfect, but it has calmed the stormy seas.

And this one:

You are right about the drop in property values causing the defaults in loans. However, one reason why the MBS was considered to be such a good investment was therte was tangible property standing behind every loan. The error was that no one believed values would drop more than 6%.

The other thing is that, at the worst, we have had only 14% of loans default. Who came up with the idea that 100% of loans would default?

Consider this math:
Let us assume at least a 10% down payment on a group of 100 houses with a price of $100,000, resulting in 100 mortgages of $90,000. If every property in that MBS loses 50% of its value, that means the properies are now worth $50,000 each. The mortgages were for $90,000 so the lender is potentially out $40,000 on each property if the loan defaults. That is a 44% loss. Meaning that, if all 100 loans default, the MBS should be worth about 56 cents on the dollar. What we know is that none of the MBSs had 100% defaults and values dropped less than 50% even in the worst markets. Let's take a case where 75% of the loans defaulted and the lender had to resell the 75 properties at $50,000 each. That is a 31% loss on the MBS meaning it should be worth 69 cents on the dollar.

It is my understanding that last year (2010) 9.85% of all mortgages were in default. Maybe there were some really awful MBSs that had 50%or higher mortgage defaults, but I don't think it is possible to make the assumption that all MBSs were in that category. Nevertheless, the mark to market program resulted in a 22 cents on the dollar valuation for all MBSs; a patently ridiculous valuation. That is why mark to market was finally dropped in February of
2009. I don't believe it is any coincidence that finacial securities have been reasonably stable since March of 2009.

The problems were caused, as you said, by:"Approximately 80% of U.S. mortgages issued to subprime borrowers were adjustable-rate mortgages.[1] After U.S. house prices peaked in mid-2006 and began their steep decline thereafter, refinancing became more difficult. As adjustable-rate mortgages began to reset at higher rates, mortgage delinquencies soared." That's from wikipedia. It needs to be said that 80% of problem mortgages were in California, Las vegas, Arizona, and Florida. Not only were people trying to buy too much house, but there was a lot of speculation going on. I was in Naples Florida at the height of the boom. There were large subdivisions of $500,00 and up homes being built. Prices were escalating and it was a common practice to buy a lot as soon as a subdivision opened, then sell the house at a profit as soon as it was completed. I had friends who had been doing that and made a lot of money. Some speculators were buying three and four homes at a time. When interest rates went up and demand cooled, it was like pushing over a row of dominoes. The government is now trying to stop the dominoes from falling to fight deflation. However, they have failed because they are unwilling to harness the profit motive. Investors will start buying the foreclosures when they see a better than average chance to profit.

The other thing that makes valuing an MBS without detailed information about what is in the portfolio is that the crisis was very geographically specific. Property values in New York City have increased after a small drop in 2009. Property values in Seattle are down only 18-20% since 2008. Those who keep screaming that the sky is falling everywhere are the ones hoping to benefit from lower prices. Great fortunes were made in the 80s when people bought packages of loans from the Resolution Trust Corporation, who sold them at just about any price to willing investors. I think there are people who are hoping for a repeat.

I pointed out that Jimmy was all wrong. He does not agree, but has graciously agreed to recant if I show up with proof that he is wrong, which puts him above the entire NY Times staff, 90% of politicians, and, quite tragically, an increasing minority of scientists.

JJ Contention 1:
All MBS were selling at 22 cents on the dollar. This is wildly off. Basically, what happened is that no one wanted to sell better MBS, so what was offered for sale was mostly the cruddy stuff. For some of the cruddy stuff, 22 cents on the dollar was way too much as it turned out. But there was never a point when all MBS lost 78% of their value.

I think part of Jimmy's misunderstanding is his failure to distinguish the underlying pool of loans from the mortgage-backed securities derived from them. Before writing any more about MBS values, Jimmy should read at least this Wiki page, which explains some of the factors involved in pricing MBS, and Freddie Mac's About MBS (pdf 13 pages), which covers some of the different MBS structures and tranching options. What any particular security sells for at the moment may have more to do with the type of payment factors incorporated in the payment flow to the different tranches than the underlying pool of loans.

Securities are priced based on expected net return over the life of the security. I quote:
Pricing a vanilla corporate bond is based on two sources of uncertainty: default risk (credit risk) and interest rate (IR) exposure.[18] The MBS adds a third risk: early redemption (prepayment). The number of homeowners in residential MBS securitizations who prepay goes up when interest rates go down. One reason for this phenomenon is that homeowners can refinance at a lower fixed interest rate.
Theoretical pricing models must take into account the link between interest rates and loan prepayment speed. Mortgage prepayments are most often made because a home is sold or because the homeowner is refinancing to a new mortgage, presumably with a lower rate or shorter term. Prepayment is classified as a risk for the MBS investor despite the fact that they receive the money, because it tends to occur when floating rates drop and the fixed income of the bond would be more valuable (negative convexity). Hence the term: prepayment risk

To compensate investors for the prepayment risk associated with these bonds, they trade at a spread to government bonds.

There are other drivers of the prepayment function (or prepayment risk), independent of the interest rate, for instance:

* Economic growth, which is correlated with increased turnover in the housing market
* Home prices inflation
* Unemployment
* Regulatory risk; if borrowing requirements or tax laws in a country change this can change the market profoundly.
* Demographic trends, and a shifting risk aversion profile, which can make fixed rate mortgages relatively more or less attractive.
Pricing MBS is extremely difficult, even if the MBS carry some sort of external guarantee. For example, holders of Ginnie Mae (entirely insured) bonds may have taken heavy losses during the last couple of years due to much more rapid prepayments as the Fed forced rates down. This is most true if they bought at exactly the wrong time, and I do believe that retail investors were somewhat victimized at the beginning of the rate drop cycle. Savvier investors, realizing what was coming, put some of the truly guaranteed stuff up for sale at precisely the time at which it would be most valuable on its face - the point at which the depth of the housing problem was not clear, but when the Fed had begun its monumental rate drop cycle:

Effectively, securities are bought and sold based on present value calculations. These present value calculations may be based on a difference between the yield on Treasuries of the expected duration and the coupon on the security. The basics of a net present value calculation can be found here in this financial calculator.

In comparing two investments over a period, you don't need the discount rate. You simply calculate the present value of the expected payments on each.

Now it should be obvious that the prepayment risk is not a smooth curve. (CPR is expressed as a proportion of the remaining principal in the pool paid off, or returned to the investors each period.) It depends on the values of the mortgages, the expected cut in rates on refinancing, and the "fixed cost" of refinancing - stuff like appraisals, etc. A very large drop in rates will always produce a big surge in refinancings, and refinancing rate will be contingent on the loan balance and the remaining term on the mortgage. If you have a mortgage balance of 30K and five years to go, you are not going to get much net return on a refinance (most of your payments are principal). But if you have a 300K loan with 27 years remaining and rates drop enough for you to get a 1% cut, spending a few thousand to refi makes absolute sense.

Thus, when rates drop as incredibly low as they have (at bottom, it was something like a forty-five year low), but underwriting standards increase, the result is that the mortgagees with higher balance mortgages who are creditworthy exit your your underlying pool very quickly. You will be left with persons with low balances and those who are not creditworthy. Even if you are insulated against the losses from the non-performing mortgages, the result is that you are going to get very rapid repayments of principal and far less interest over the term than you expected.

So now you have your principal back, but you originally paid a high price for this security precisely because you wanted that rate, and your investment options for your returned principal are much worse than they would have been if you had just put the money in a CD or in treasuries.

For example, let's assume that you had 100K to invest in the first two months of 2007. You could have put it in a five year bank CD at about 5.15%, insured in February. No risk of loss, but you knew you would have to face reinvesting your money in five years. You could also have bought into some Ginnie Mae pools with substantially higher coupon rates, although you might well have had to pay a premium to get them. Scroll down for February values. An innocent might be quite happy to pay a premium to get a guaranteed return of principal and a theoretical 6% rate. However, when prepayment rates picked up hugely over the last few years, you might now have gotten 70 to 90K of your principal back and be faced looking for a way to invest it. Thus, you overpaid, and you took a loss compared to other investments you could have made. You could, for example, have bought a ten year treasury in February of 2007 that yielded between 4.75 and 4.8%.

All three of these theoretical investments have the same backing - the US government. Thus the credit default risk (if any, implicitly zero) term is the same. However rate risk and, for the MBS, the duration (prepayment) risk is quite different. At the time you buy MBS securities, you make a number of assumptions, and if they are wrong, you may take a loss even if the underlying loans do not deliver a loss. One nasty way of taking a loss is to buy MBS with a lower coupon rate and then find that your duration is much longer than expected, so that in the later years of the security you are losing a lot of return compared to having your principal back and being able to reinvest it.

This post is pretty long already, and I have not yet covered most of the factors in pricing MBS. I think I will end this here, and pick up when I have time. One aspect of this I would like Jimmy to accept is that the primary determinant of market value of an security may have nothing to do with underlying loss expectations on the loans.


I generally skim your longer posts first then come right back to it for a more careful read. You have already won me over with this though.

"He does not agree, but has graciously agreed to recant if I show up with proof that he is wrong, which puts him above the entire NY Times staff, 90% of politicians, and, quite tragically, an increasing minority of scientists."

LOL! Indeed it does!

Okay, I'm now sitting back comfortably in my chair making a second more detailed read of your analysis. :)
I have read it in detail now and I have an additional thought to share.

"Pricing MBS is extremely difficult, even if the MBS carry some sort of external guarantee."

"One nasty way of taking a loss is to buy MBS with a lower coupon rate and then find that your duration is much longer than expected, so that in the later years of the security you are losing a lot of return compared to having your principal back and being able to reinvest it."

I am spending a great deal of time thinking about something similar right now. It is my plan to put my entire IRA into one 30-Year TIPS bond (in February's auction). You'd think it would be an easy decision. It pays about 1.9% over inflation for the next 30 years. I don't need the money for about 30 years.

It isn't that easy though.

I have to worry about possible US defaults in the future.

I have to worry that my IRA might be confiscated at some point and that TIPS bond might therefore be sold prematurely (at extremely unfavorable prices perhaps).

I have to worry about changes in taxation rules (maybe IRAs lose their tax deferred status or maybe tax rates go so high that investors therefore require much higher real yields to compensate).

Any one of these things could and would alter my present decision to buy, perhaps significantly.
Last year was kinda a bizarre year for the mortgage market. In the first half of the year, you had a decent number of home sales keeping mortgages for purchases stable, thanks to the home buyer credit. In the second half of the year, that changed as demand crumbled when the credit was withdrawn. At the same time, you had very low mortgage interest rates throughout much of the year cause a mini-refinancing boom. 2011 will look very different, as the housing demand continues to struggle and mortgage interest rates have begun rising.
It's not looking good for you Jimmy. 2000 words and she hasn't even gotten to tranches. ;-)

As a real-life example, one of the NPR programs bought an MBS for 1 cent on the dollar and still over-paid. It returned only 0.45 cents, as in half a cent, on the dollar.
I haven't even gotten to loan losses yet!!
Mark - I have come to think of it as The Great Discontinuity.

Will we get our heads out of our navels and deal with reality, or are we Italy by 2020? I think that would be the worst possible result, but to date we haven't shown a collective ability to deal with our situation.

It certainly makes buying such investments challenging.

"It certainly makes buying such investments challenging."

It makes buying any investments challenging. Sigh.

"The financial policy of the welfare state requires that there be no way for the owners of wealth to protect themselves." - Alan Greenspan (1966)

In hindsight, 1966-1981 was an awful era.

My main concern is that we've tried to combine the deflationary Great Depression with the inflationary 1970s. I just don't believe that two wrongs will make a right (at least over the long-term).
Jimmy J vs M_O_M brings to mind the great battle of wits between the masked man and Vizzini; someone should have warned Jimmy: Never go up against M_O_M when death is on the line...
Shrinkwrapped, I think you've got that backwards: Westley, after all, lived. It was Vizzini who died, and for the same reason: he did not understand an unknown unknown (Westley's immunity to iocane powder in Vizzini's case, the factors that go into valuing a MBS in Jimmy's case).
Gosh, I'm late to my own funeral. How bad is that? LOLs.

I get it that not all MBSs were at 22 cents on the dollar, but what was happening to the banks and other holders of the MBSs was that the shorts were putting out the story that that was the case and it was working. The equity values plunged, their holdings of MBSs(in my understanding) were all marked to market at 22 cents, so their capital base shrunk below the mandatory required level. As a result, there was no way for them to raise new capital. The paranoia got so bad that banks wouldn't even lend to one another in the overnite market. So, it is my contention that the crisis was brought on by the mark to market rule imposed by FASB and the SEC and the shorts (and who knows who else - George Soros maybe?) talking the values down. We don't know what the values are, but it seems to me (and I could be wrong) that since mark to market was lifted new confidence has been created and that for the majority of the MBSs the value is well north of 22 cents. Most of my information has been gleaned from reading Steve Forbes in his magazine. He is not unsophisticated, but I admit he could be wrong.

Here's a link to an article about valuing MBSs that supports my thesis. Admittedly, USA Today is not Barrons, Forbes, or WSJ, but the article made sense to me.

I've researched other documents and now realize that Wall Street divided up some of the MBS tranches into interest only (IO) and principal only (PO) as well as slicing them according to quality. The math is way beyond my meager level of comprehension, but I can see where IO tranches could go to a very low level as there is no property standing behind the security. The same applies to sub-prime tranches, but the entire loan market was, at its peak, 20% sub-prime.

My point, and maybe I didn't make this well enough, is that when you look at the entire MBS market, you cannot mark the whole market to the only bids they could get at the peak of the crisis. I accept that I could be wrong there too because I was relying on Steve Forbes opinion. Maybe I should quit believing that he knows whereof he speaks.

It's late and I'll have more time to study the links tomorrow and see where else I'm wrong. I appreciate your taking all the time to educate me.
I'm back. After looking closely at the post, I realize where I'm going wrong. You are thinking in terms of the relative value of one investment over another. Whereas I'm thinking only in terms of what is necessary to calm the waters in the hybrid securites market and the financial system as a whole.

I can agree that very few were willing to put any money up to buy an MBS at the peak of the crisis -irregardless of the properties in the MBS. And one of the main reasons was they were opaque. If I want to buy a closed end bond fund, I can check out the portfoloio and make a judgment about the risk. With the MBSs that was not possible. And if I'm considering where to invest my money at the peak of the crisis, I'm going to opt for Treasuries over an unknown quantity. That is in fact why the TARP program never happened. Even though Treasury was willing to bribe investors to take the risk, so few were willing that the whole idea fell apart.

Another factor in the low and very few bids they were able to get was the question of who actually owned the mortgages inside the MBS. Another major unknown. That was kicked off when an MBS owner from Germany tried to foreclose on a delinquent mortgage in their MBS that was in Cleveland. Some judge rulled that they were not able to proceed and that created a furor. I see that issue has just come up again. Many of the foreclosures are being challenged on the basis that there is a question of who actually owns the mortgages. Thsi may be related to the principal only, interest only provisions of some MBSs.

Securitization of mortgages is not new, but it used to be orderly. For years one could invest in GNMA funds for yield, and though the risk of lower yields from early payoffs was there, they were preferred by many over corporate bonds. It appears that when Wall Street got involved, things changed for the worse. They became more complex, riskier, and corners were cut that has put us in the problem we have today.

All that said. (I've been bobbing and weaving, trying to avoid the KO by MOM.) I do agree with your statement: "One aspect of this I would like Jimmy to accept is that the primary determinant of market value of an security may have nothing to do with underlying loss expectations on the loans." Recantation is in the air!
The Fed bought MBSs eventually.
Charles said, "The Fed bought MBSs eventually."

And was that a good thing or bad thing?

The other option was setting up a "bad bank" or a Resolution Trust Corp. II and putting all the MBSs in there, then working them out over time. My impression is that that idea was too unwieldy or was thought to be too much government involvement. Since the FED is less transparent and has unlimited resources, they could buy enough of the MBSs to provide liquidity to the banks and then gradually resell them back into the system over time. Many don't like this idea. Ron Paul is one who comes to mind. But no one has come up with another solution that won't collapse the system. At least that is my view.

All the FED is doing is playing for time, as they are sure the economy will come back. When the economy comes back most of the present problems will be solved. Of course, the flaw in that strategy is that the Obama administration does not know how get out of the way and let the economy recover. So, in my view, we are in uncharted territory.
Jimmy, I brought up the Fed purchases because it was TARP in sheep's clothing.

Since you asked I'll give you my opinion. Whether the Fed's actions were good or bad depends on your position in the economy. But for the economy at large the Fed's actions were bad because they encourage malinvestment.

Malinvestment is always encouraged whenever the investor is not on the hook for all of the investment's losses. We're going to be wiping out all the recent modest gains with increased state and local taxation in order to prop up government employee malinvestment.
Jimmy - when I recover from my Medicare wage receipt-induced whimpering, I will get back to this.

Note that the article you linked gives a price of 22 cents not for all MBS, but a block of CDO's owned by Goldman Sachs.

CDOs are not the same as MBS. CDOs(Collateralized Debt Obligations) could be formed largely from the lower tranches of MBS. When you have an easy money environment, those lower tranches of MBS will generally outperform ratings. But when the environment reverses, that is where the losses concentrate. My hunch is that those who paid 22 cents probably got burned. Goldman was into the sleazy loan business.

I will do this in order. Next up, loan pool losses. Then MBS types and tranches. And now I have to add CDOs.
I was watchingh CNBC this morning and they had Godlman Sachs head of Financial Instruments Trading department on. I listened very closely to what he was saying and, not surprisingly (at least to me) he said the MBSs were doing well, that most were holding up and did not have as much bad paper in them as every one assumes. I realize he is talking his book, but there may be a smidgin of truth in there.

At the point where the Treasury and the FED acted to prop up the capital bases of the financial institutions, IMO, we were way beyond the issue of moral hazard or encouragement of malinvestment. They were, and still are, engaged in keeping the system from coming apart at the seams.

Last nignt I spent some time on the Volokh Conspiracy website where they have a post on the meaning of the MA ruling that the chain of ownership of many mortgages is broken. It appears that there were many mistakes made in transferring ownership rights to the mortgages or the PO or IO rights as well.

The comments have a lot of meat in them. The one that frightens me is that the real estate market may become a "cash only" market. Talk about something that would destroy real estate prices!
You can read it here:
Jimmy - that's not going to happen.

The mistakes that were made do not endanger mortgages well handled. The MA case was an old one, and procedures should have been tightened up already.

Among the comedy of errors in that case, they appear not to have actually assigned the mortgages until after they started the foreclosure. The lower court judge took umbrage (this meant that they filed a court case under the wrong names). He reasoned that they did not have to have recorded the assignment, but that they certainly should have done the paperwork that would allow the entity attempting to foreclose to legally do so.

The MA Supreme Court followed that line of reasoning.

This case in no way threatens properly handled mortgages, and even if the mortgage wasn't properly handled before, all it means is that the entity foreclosing should do its paperwork before trying to foreclose. I heartily agree.
Jimmy J said:

At the point where the Treasury and the FED acted to prop up the capital bases of the financial institutions, IMO, we were way beyond the issue of moral hazard or encouragement of malinvestment. They were, and still are, engaged in keeping the system from coming apart at the seams.

No, the Treasury and the Fed were involved in keeping the current players in the current financial system from coming apart. We need a financial system. We don't need Goldman Sachs, JPM, BoA and other such swine.

Also, for some detailed reading on tranching see the much-missed Tanta here, including this picture.

For CDOs see here, and most especially this picture.

The AAA tranch of the CDOs are the BBB- or lower tranches of the original MBS, elevated through the "magic" of bad math and captive ratings agencies!

Then remember that most of the loans were crap and loss rates are or will be much higher than the "ass-umptions" used.

2 years after the TARP debate, people still don't get the difference between the tranches of securities, and the loans and houses behind them.

(M_O_M: Carl Frank never did understand this - or at least he wouldn't admit to it, since Goldman-Sachs are his heroes. His undying pro-Bankster bias is why I no longer read him.)
Post a Comment

Links to this post:

Create a Link

<< Home

This page is powered by Blogger. Isn't yours?