Laurence Lewis has a rec list diary up today about a federal lawsuit against "the banks" over misleading mortgages.
The response and debate have been interesting. Some people seem pleased that the federal government is "finally" going after the banks, while other seem skeptical and wonder why this suit took so long to file.
I think it's safe to say that there is a strong current in the comments that the case for suing and prosecuting the banks is "obvious," a slam dunk, and so there is puzzlement that it took three years to get this case filed. (Btw, I don't think it's accurate to portray this filing as though the federal government is finally taking action, because as the NY Times article clearly states, subpoena's were issued over a year ago.)
I think it's also fair to say that the comments reflect that there is a lot of unfamiliarity with just what went wrong with the mortgage securitization markets to cause the crash of 2008, and just what the banks did that was "wrong." But here are a few random observations about why these cases are not so obvious from a legal perspective, and why I'm somewhat pessimistic that the federal government will win all of them in the way many of you want. I should say upfront that many years ago I used to work in finance. I actually specialized for a few years in mortgage backed securities, and I think I understand some of what happened in 2007 and 2008. I hesitate to explain it because for many people explaining is considered excusing.
1. There wasn't one big fraudulent scheme, despite what you've read on DK
So many news stories about the mortgage mess have used the term "fraud" that many people seem to have concluded that there was one huge fraud scheme or conspiracy of a few big banks to rob the public.
The problem is that there was more like a collective madness that involved people engaging in arguably many different kinds of fraudulent activity, from the big banks that sold mortgage backed securities to investors, to the mortgage banks that made loans with little or no paperwork or proof the homeowner could pay the mortgage, to even some property owners.
In a way, the existence of widely dispersed fraudulent activities by many actors, at many steps in the creation of the toxic mortgage backed securities that tanked the financial system and economy, makes a federal case really difficult. For example, Bear Stearns created some of the crappiest collateralized debt obligations (CDOs) that were issued and that screwed up the system. Arguably, when they issued those CDOs and told the investors they were triple A rated, they lied, which would be securities fraud.
Arguably, but not indisputably.
That's because CDOs were packages of other banks' mortgage backed securities (MBS) and Bear Stearns' assertion that its CDOs were really safe depended on the assertions by other banks that their MBS were safe. If you were to try to sue Bear Stearns, or what's left of it, it could always say, we were lied to by Lehman Brothers when they sold us their crappy MBS.
If the underlying Lehman Brothers MBS were indeed crappy because the mortgages backing their MBS were crappy, and you try to sue Lehman, Lehman could say, Countrywide lied to us about the quality of mortgages. To top it off, you had the ratings agencies like S&P and Moody's, supposedly the smartest guys around, running their super smart math models telling everyone that the securities were safe.
In each case, holding a player accountable is more likely going to depend on whether that player did "due diligence" than on that player deliberately lying. Due diligence means a "diligent" investigation to the extent that is "due," of the quality of the assets that backed their securities. But if they were actively lied to, does that mean that their own statements were lies? If they tried to investigate, but the subject of their investigation lied to them, did they themselves lie?
Ultimately, the best bet would be a very massive investigation and suit that showed that various players colluded to lie to each other so that they would have no excuses about the inaccuracy of their statements to the public -- a not unlikely possibility, considering that the ratings agencies got paid to give good ratings, and changed some of the their business practices to get ever increasing amounts of rating work for MBS and CDOs. But that could take years to prove. It will involve "discovery" a preliminary investigation in a lawsuit, that will mean reading literally millions of pages of documents, taking informal testimony (depositions) from hundreds of executives, and hoping to find the smoking gun or guns. Problem is that the gun may not be there. Everyone might actually have become extremely reckless and greedy and believed their own advertising and math models.
2. What part of the Federal government is suing, anyway?
First of all, most of the comments show that not everyone is familiar with who is suing or why. The federal government is a very big institution and lots of its branches and organs had some sort of jurisdiction over the things that went wrong in the mortgage market -- including the Securities and Exchange Commission, the Department of Justice and various banking regulators. State attorneys general also have jurisdiction over certain aspects of the mortgage mess at the level of making mortgages under state law.
The fact that some folks in the comments to Laurence's diary are wondering whether criminal charges can be thrown in shows that many folks don't realize that this suit is not being filed as a criminal matter by the Justice Department. There is no possibility that this particular lawsuit will involve criminal charges. That's because this is not a criminal case and the branch of the federal government suing is not a law enforcement agency. Many different parts of government at different levels are indeed investigating how to sue or prosecute over issues raised by the financial crisis, but it's going to take a long time, and there are going to be lots of fights over how to coordinate those efforts, as we've already seen.
This suit is being filed on behalf of Fannie Mae and Freddie Mac, the two big semi governmental issuers of mortgage backed securities and guarantors of home mortgage loans. When they became insolvent, they were taken into "conservatorship," a sort of special bankruptcy trustee situation, by the Federal Housing Finance Agency (FHFA). Its job, among other things, is to conserve Fannie and Freddie's assets. One way of doing that is to sue people who ripped them off.
That is what this case is about. It is not about the federal government "in general" going after "the banks." It is about the Conservator of Fannie and Freddie going after the investment banks that packaged mortgages and sold mortgage backed securities. It is not, for example, about the more local retail banks that made bad mortgages to home buyers.
One great thing about this suit, however, is that Fannie and Freddie are suing almost like normal investors -- saying they were ripped off over the quality of the MBS they bought. If they win, it opens the floodgates to other lawsuits by private parties. But weirdly, a large proportion of the parties that will be able to sue under Freddie's and Fannie's model will be other banks. That's because overwhelmingly, the purchasers of MBS are other financial institutions, not regular folks.
3. How difficult is the case being made by this branch of the federal government?
The theory behind the case seems to be, as described by the NY Times:
The suits will argue the banks, which assembled the mortgages and marketed them as securities to investors, failed to perform the due diligence required under securities law and missed evidence that borrowers’ incomes were inflated or falsified.
In other words, Fannie and Freddie bought mortgage backed securities from, say, Lehman. Lehman bought mortgages from other banks and mortgage companies. Lehman did not give mortgage loans. It was not a retail bank that dealt with home buying customers. It had to buy these mortgages from other players, and those players would have provided paper work showing whether the mortgages were good credits and safe, or not. Lehman then transferred these pools of mortgages to trusts. These trusts would have names like, "Lehman Mortgage Backed Security Trust 2006-M." The trusts would then issue trust certificates to investors. The trusts promised to "pass through" all the money it received from home buyers to the holders of the trust certificates. However, when too many of the home buyers defaulted on their mortgages, the trusts weren't able to pay the holders of the trust certificates, which meant the trust certificates or MBS became worthless
In order to sell the MBS trust certificates, Lehman had to write and issue a document called a prospectus. Except in limited circumstances, in the US a security cannot be sold without a prospectus, a document that describes the security and the risks involved in investing in it, in excruciating detail. A prospectus for an MBS might be around 200 pages, and supporting exhibits another 2,000 pages. (Think of it as something like the list of ingredients that must accompany most packaged foods. It is also why whenever you hear a commercial on TV for an investment, there is a statement at the end saying that this is not an offer to sell a security, which can only be done with a prospectus.)
If there is a lie in the prospectus, then Lehman has committed securities fraud and is liable to the investor who loses money as a result. The lie doesn't have to be deliberate. If Lehman was lazy in carrying out its investigation into the quality of the mortgages it bought, then it didn't carry out due diligence. They weren't allowed to have been willfully ignorant of how crappy the mortgages were. But Lehman can't be sued successfully simply because it made a statement about the housing market that was true when the security was issued in 2006 but that wasn't true in 2008, or that failed to predict the future. The explanation from the NY Times suggests that the FHFA acknowledges it is on shaky ground; it doesn't say that the banks deliberately lied in their prospectuses; it says that they "missed evidence that borrowers’ incomes were inflated or falsified." But they didn't necessarily have that evidence; the banks making the actual mortgages had that evidence. So it all depends on whether the MBS issuing bank's due diligence was so sloppy that a diligent banker would have found the missing evidence.
This is where the problem of math replacing due diligence comes in.
The problem is that math really took over the work of due diligence. Way back in the olden days when I was doing this stuff, when mortgages were packaged, some poor schmucks at the bottom of the hierarchy (ehr, in other words, me) were sent to some document warehouse in New Jersey for a week to read through a room full mortgage documentation and then write a 100 to 200 page report. That was the due diligence. Even if it turned out that something ended up not being true in the prospectus, it wasn't because we didn't do our damndest to find out. The defense to a statement not being true was a sincere effort to get to the truth -- ie a due diligence investigation.
To be honest, I have to admit, that if the bank that made the mortgages had made crappy mortgages and created fraudulent mortgage documents, despite spending a week in the document warehouse in New Jersey reading those documents, I doubt I could have discovered the fraud. How would I know that the mortgage documents I was reading made by a crooked loan officer in a bank branch in Delaware were fraudulent?
As I was leaving the field, however, the pools of mortgages being packaged were getting bigger and bigger. They went from being pools of $60 million, for example, which was just a few hundred mortgages, often made by the same bank or housing developer, to $100 million, then $500 million to billions. I read a really crappy Bear Stearns CDO prospectus summary for $1.5 billion. It was impossible to do due diligence for such pools.
But at the time, the banks started hiring math PhDs and computer programmers to replace credit analysts and junior accountants and junior lawyers to do due diligence. These math geniuses came up with a novel idea: You didn't really need to look at every mortgage; you only had to look at what, statistically, had happened to mortgages in the past. In other words, random samples and historical data replaced real due diligence.
This was not as crazy as it sounds. We rely on statistics and probability a lot to predict events. For example, every day, DK has a round up of polling data on elected officials and candidates. It was really at mid century that the mathematicians began to prove to the business and political and medical worlds that very small samples provided a remarkably accurate picture of much larger groups. Pollsters can now use a few thousand telephone calls to predict with very high accuracy how 120 million people will vote. This was the theory behind the new approach to figuring out how safe mortgages were and in turn how safe MBS were. Do random samples (or maybe don't do even random samples), and just study what the historical default rate was. It it has been the same for decades, there's a pretty good chance it will be the same in the future.
So the question is, are statements based on probability fraudulent? If Lehman said, for the last 50 years the default rate on mortgages is X percent, and we've built a safety factor into the MBS that assumes that X percent can default and your MBS will still be perfectly safe if the default rate is X percent, and then hurricane George Bush, the worst president in American history, runs the economy into the ground, and the default rate skyrockets to 10X percent, has Lehman engaged in fraud? This is going to be a tougher case than many people realize. Will the lawyers and judges (and juries!) be able to understand the math models?
Also the case is going to turn on whether the math geniuses realized or should have realized that the explosion of the MBS market and the feedback of the MBS market itself into the local real estate markets of this ocean of loan money itself made the historical data inaccurate, such that relying on it was a kind of fraud.
I of course think that what happened was reckless and forseeable. The problem is, whether under the Securities Act, the federal government can prove it against an army of corporate finance lawyers. Frankly, I'm not optimistic. To me it makes sense that this is taking a long time because it's a very difficult case to make. The SEC under the Obama administration has been studying this for a long time as well.
4. Credit Enhancement went poof - or "System Risk".
I think that closely related to the problem of whether the math models being wrong can be proven in a court of law as being fraud, is the problem of whether it was fraud not to be able to predict the unpredictable effects of a total world historical financial shit storm.
One way of looking at this is to look at the problem of "credit enhancement." Some MBS and CDOs had various kinds of credit enhancement -- in other words, various guarantees.
It is fundamental to analyzing a credit risk of a guaranteed debt taken out by a risky borrower but guaranteed by another party, that the rating of the debt is based on the guarantor.
Let's say you have a great credit rating, but your brother in law has a terrible credit rating. He asks you to co-sign a car loan. The car dealer doesn't really care about your brother in law's credit rating because you are equally on the hook as he is. The guarantor's credit is what matters.
But what happens if suddenly you, the party with the great credit get slammed financially?
That's kind of what happened in 2008. Many MBS had credit enhancements built in -- some entity would come to the rescue if things got really bad. So let's say that Lehman issues a MBS with some shaky mortgages. But it buys some sort of insurance or guarantee from Bear Stearns to stuff into the MBS trust -- well then the shakiness of the mortgages doesn't matter. All that matters is Bear Stearns, right?
But in 2008, all these investment banks, like Bear Stearns collapsed at the same time, which meant all of their cross guarantees became worthless, which meant that all the MBS were suddenly crap, which meant that banks' balance sheets were wiped out, which meant that banks credit ratings were crap, which meant that their credit enhancements were crap, which meant that even more MBS became crap, which meant that more bank balance sheets were wiped out, which meant -- you get the picture. This is what was referred to as Systemic Risk -- the risk to the entire system of certain banks failing.
That really was the impetus for making policy on the basis of some banks being "too big to fail" -- not just that they were rich and powerful and could extort a bailout. It was that their interlocking guarantees, collateral, credit default swaps, and so forth were threatening to bring down the entire system. It was as though a hurricane were approaching a city and at just that time, all the property insurance companies collapsed.
5. Who is left to sue?
I'm always puzzled by left blogosphere claims that the "banksters" tanked the economy and lost nothing and just got bonuses.
In fact, so many of the investment banks that issued MBS and CDOs went bankrupt, that my biggest worry is that there are very few remaining financial institutions to sue and to recover assets from -- compared to the extent of the damage done. It's sort of like one bankruptcy trustee suing another bankruptcy trustee. Lehman, Bear Stearns, Merrill Lynch, are all gone. Goldman Sachs is pretty much the only traditional Wall Street underwriting investment bank that was left standing.
While it would be great to see justice done and fraud exposed, I don't see that there are going to be a lot of defendants to collect damages from.
NB: Edited for clarity.