Mathematicians on Mortgages

Jun 09 2008 Published by under Mathematics

In a nice little coincidence, recently two mathematics bloggers have decided to give a bit of a description of the subprime mortgage market crisis.  Neither is an economist, but that's probably okay, even preferable, considering it was the economists who got us into this mess.

Ben Allen of Plektix has written a two-part discussion here and here, and

Mark Chu-Carroll of Good Math, Bad Math has written a two-part discussion here and here.

All of these articles are worth a read!

6 responses so far

  • Babs67 aka the fiancee says:

    Ben Allen did a good job illustrating the mortgage-backed securities process. It's too bad he wasn't able to provide that lesson to the bankers who got us into that mess in the first place.

  • Babs67 wrote: "It’s too bad he wasn’t able to provide that lesson to the bankers who got us into that mess in the first place."

    This seems to be an example of 'tunnel vision': each banker in the business was focused on their little piece of the puzzle, and nobody was looking at the big picture. If they had taken a broader view, they would have seen that the system as being managed could not sustain itself.

  • Personal Demon says:

    I enjoyed reading all of these articles. What I found most interesting from both authors was that both authors demonstrated a mathematical basis for how complex systems become "blame free" systems. I think it sheds a lot of light on how a group of well-intentioned, well-organized people can do unfathomably awful things.

  • PD wrote: "What I found most interesting from both authors was that both authors demonstrated a mathematical basis for how complex systems become “blame free” systems."

    I've been saying for years that a good definition of 'bureaucracy' is a system which is designed to eliminate blame.

  • Babs67 aka the fiancee says:

    PD - one small flaw in your comment about "well-intentioned" people. A friend of mine worked for HSBC, one of the big offenders in the sub-prime mess. She worked with the team that developed the risk and profitiability models. The sales people kept pushing the team to open up the models to allow them to sell more loans. The team began building models to look at these fringe populations (high-risk). Since there was no previous history with these groups, the assumptions that they used in those models was based on dissimilar populations. For example, the model had to do a "what if" scenario on rising rates. So, they included the assumption that the attrition and default for these populations would be the same as average risk populations. The assumption was that if rates rose and payments increased, these people would refinance into a new mortgage. That assumption was inherently flawed. They were offering mortgages to people who couldn't qualify for any other type of mortgage so if they couldn't make their payment, there were no other options for them except to default. My friend questioned them, but they were under so much pressure to open the flood gates and make the short term profits that they went against common sense. Management also probably assumed these mortages would be sold off and if and when they defaulted, they would be someone else's problem by then. Greed, plain and simple also contributed to the ultimate demise of this system.

  • Babs67: You make a good anecdotal argument as to why complex societal systems need some sort of regulatory mechanism. It is aesthetically appealing to imagine that a completely 'free market' will follow an evolutionary-type course and solve all its own problems, but people seem to forget that evolution is filled with dead ends. If everyone just follows their own self-interest with no regard to a 'big picture' the system can steer itself off a cliff.