Last weekend I posted about six flaws with the $1 trillion toxic waste cleanup plan. Yesterday, I added a seventh flaw -- hedge funds and private equity firms are worried that if they participate in the program, political pressure will strip them of their profits.
And now, a bigger problem has reared its ugly head -- investors are becoming less eager to buy the U.S. debt needed to finance all these bailout programs. This puts on even more pressure to come up with a plan that works.
So what to do? Pay $70 million to get hard data on the value of toxic waste rather than putting hundreds of billions at risk to play a game that might get the toxic waste off banks' books. To explain why this $70 million plan to get data will work, let's examine one of the biggest problems with the $1 trillion toxic waste removal plan -- there is no agreement on how to price it.
Banks have booked it at 60 cents on the dollar but investors are only willing to pay five cents. Pay less than 60 and banks take a loss and need to raise more capital; pay more than 60 and taxpayers and investors lose.
To explain this pricing problem it's important to recognize that mortgage-backed securities (MBSs) are backed by individual mortgages, say, 4,000 of them in a typical bundle. The reason that nobody knows what such toxic waste is worth is that there is no information available in one place about the likely future cash flows of each of those 4,000 mortgages. If 15% of those mortgages, or 600, are in default and the other 3,400 are likely to keep paying the mortgages, then it becomes possible to put a specific value on that MBS.
But that information does not exist so the U.S. wants to finance an auction process that is designed to set a price on the toxic waste. But rather than put hundreds of billions of taxpayer money at risk, I think the government would be better off hiring people to examine each individual mortgage and draw a conclusion about the value of the MBSs based on hard data.
How much would it cost to hire such data crunchers? A colleague ran some numbers and concluded it would be much less expensive to get this information than to simply take a loss due to the lack of data. He estimates that it would cost $28,000 to price 4,000 mortgages in the example I used above.
He based this on the assumption that an analyst who is paid $40/ hour could estimate the cash flows of six mortgages per hour -- yielding a cost per mortgage of about $7. With the expected loss on a single mortgage at 40%, a $300,000 mortgage would generate a loss of $120,000. So the value of the information would be far lower than the benefits from getting the information.
If there are 10 million mortgages buried inside all that toxic waste, the total cost of putting a hard value on those mortgages would be a relatively paltry $70 million. Getting this hard data on the value of the toxic waste would not solve the problem. It would merely make it easier to price it accurately. And those prices would take much of the uncertainty out of the price setting process. As I posted, the FDIC could then buy the defaulted mortgages out of the MBSs and restructure them.
And then the toxic waste could trade on the open market since the defaulted mortgages would have been stripped out of them. With the cost of financing all these bailouts going up, we need to find smarter ways to solve the problem with less taxpayer money.
I think my $70 million data plan would get things headed in the right direction.
Peter Cohan is president of Peter S. Cohan & Associates. He also teaches management at Babson College. His eighth book is You Can't Order Change: Lessons from Jim McNerney's Turnaround at Boeing.
Will Rich Ross really have a positive impact on Disney's movie operations?
Consumers sick of recession, may spend extra on holidays
Corrections, seasonality and Santa Claus
Reader Comments (Page 1 of 1)
3-26-2009 @ 12:02PM
minhong220 said...
This is ridiculous. $40/hr analyst reviewing a single mortgage in 10 minutes? That's the stupidest idea yet. A finance analyst earning $40/hr is basically someone with 0-2 years experience (~$30/hr in salary = $60,000 yr and ~$10/hr going towards benefits), and he's supposed to estimate the likelihood of a default on a mortgage (i.e. verifying income, property values, household debt levels, refinancing options, etc...) in 10 minutes? Or is this actually saying that the average default rate should just be set at 40% for all MBS's? If so, what's the point of looking at each mortgage individually?
3-26-2009 @ 12:40PM
Iridium said...
The problem is that we can't even find out who holds the mortage payment. Essentially the banks are getting phantom payments.
We can find the mortgages that are still being paid but there is almost no way to track them to the actual person making the payment. The only thing you can look at is payment history. You won't be able to assess the risk of default on any one of the mortages in the security.
That is what makes the asset toxic. You have no way of knowing how many of the current mortgages will default. Also the value of all the mortgages is far higher than the actual worth of the property. Even if someone purchases a MBS at $.60 on the dollar the actual worth of the property that backs the paper might only be $.50 on the dollar.
There truly is no way out of this situation other than letting the banks fail.
3-26-2009 @ 3:34PM
Thomas Paine said...
"We can find the mortgages that are still being paid but there is almost no way to track them to the actual person making the payment. The only thing you can look at is payment history."
That is contradictory to logic. If a TRACE can link an actual, individual mortgage within a MBS to its payment history, then enough of a link exists to go in either direction.
From MBS to Individual Mortgage, to Mortgage Holder, to Mortgage Payer. If any MBS instrument cannot be "opened" in such a way as to reveal the individual mortgages, only THEN would there be a problem.
Otherwise, open them and follow the TRACE all the way back to the payer's mortgage-details. Run an analysis on a couple-few dozen elements - geography, housing values w/in that geography, income level, job verification, etc.
It wouldn't require the actions of a financial analyst, not for the actual number-crunching. We have these wonderful inventions of the 21st Century to handle that task : Computers.
What would be required is for some of the best financial minds to develop, in conjunction with software engineer gurus, the programs that would do the statistical and data analysis needed.
What Peter is proposing should be considered a Given. Of course the actual dollar value should be determined! In our everyday world, no one would buy *anything* without knowing its true worth. Why should this be any different?
Buying a "pig in a poke" hasn't come back in fashion, has it? I realize that the current situation has many people running amok, and equally gullible, but can't we all agree to spend a few dollars to save a few billion $$$$?!
Excellent point, Peter.
3-26-2009 @ 5:05PM
Kevin D said...
Alot of this is really hard to understand for your average or novice investor.
I took out a personal loan a while back and was supposed to be offered a certain APR but that APR was set higher in the finalized contract and when I complained I was informed there was nothing the bank could do since my debt had already been sold to the "market" presumably in one of these cobbled together asset-securities, so the bank was really just the middleman in all of this. I would assume that these MBSs worked the same way so how is it that these banks who have sold the MBSs to wall-street thusly receiving the full market value at that time now have an issue of needing to receive a "fair" value for these said securities?
It's all complex, but it looks like all of us little guys who have taken a bath with our investments are going to be paying twice the value for these securities to prevent a "loss" by these banks. I guess that's "fair".