Quant models. Forgotten variables?

September 14, 2009

Here is a widespread claim: The problem with quant models is that they forgot a few variables.  This position on the causes of crisis plays into two dominant arguments: The first is that the methodology of modeling is fundamentally sound; what was wrong was the models themselves which need tinkering. The second, also discussed in this article in today’s NYTimes, is that social network analysis is part of the solution because it can track the contagious spread of ideas about markets that can lead to panic that exacerbates market malfunction.

My question is: what is the methodological basis of social network analysis?  It is perhaps ironic that some of the biggest disciplinary critics of the economists are now working on an algorithmic platform that allows them to be folded into the logic of economics and quantitative finance. The Times article cites an NSF grant awarded to at team at Cornell to work on collective behavior. The team is populated by computer scientists, economists and finance experts… with narry a sociologist in sight.

This article by Bruno Latour explaining how new quantitative techniques that allow individuals to be tracked, rather than treated in aggregates, are fundamentally different than traditional statistical tools, may be useful starting point for addressing these issues. Latour’s insight is that the history of quantification is not just a story about doing better statistics by adding factors to combat incompleteness; rather the progress of models has involved the rise of novel type of quantification methods that have changed the nature of the world that these techniques presume to model.

Will quant models survive?  Yes.  But the simple narrative that financiers will continue to tinker with their toys might not be enough to explain their resilience…

10 Responses to “Quant models. Forgotten variables?”

  1. Chris Jefferis Says:

    This is a fascinating topic Martha. Here is a thought experiment – take a mortgage backed security. You have explored how an MBS has a risk rating calculated on the basis of the use of FICO scores of mortgage holders i.e. an aggregation of past expenditure and a highly mutable quantification of risk. Imagine an MBS based on real time data about the income and expenditure of mortgage holders. It’s not to different to what CFO’s do in a corporation. The difference is that the corporate structure gives them a right to monitor the expenditure of employees to ensure that the companies finances and therefore the value of the companies shares have some integrity, whereas privacy legislation rightly prevents financial institutions from monitoring the lives and expenditure of mortgage holders to ensure that the MBS has integrity.

    • marthapoon Says:

      This is interesting Chris. I’m not sure that privacy laws would prevent the idea you’re floating. Let’s suppose that at the point of origination the borrower were to sign a document permitting the appropriate agent to track their FICO scores. Indeed, there would be a way of generating a fluctuating assessment of MBS value througout the life of the mortgage. (Even better to track shifting home prices as well).

      What tickles me about this idea is that it demands a fundamental re thinking of creditworthiness. In US housing, despite the introduction of scores, the rhetoric tends to conflate risk scores with an old fashioned notion of creditworthiness as ‘willingness to pay’. Yet under scores, the credit rating has become a perpetually shifting condition of the person. Likewise, presuming that loan assessment need only take place once, at the point of origination, is certainly a hang over from a time when it was too expensive to track loan quality.

      …To take your point a step further, if interest rates are justified by risk vs. return, the interest rate on the mortgage should also fluctuate as scores move. The ultimate self regulating cybernetic mortgage market! Imaging these kinds of logical extensions of the system in place are fun. But of course, this is a very different thing from understanding the history of how this system has come into being…

  2. vimothy Says:

    OT but I recently saw some commentary on your site about Gillian Tett’s book and Donald MacKenzie’s review. There seemed to be some odd arguments going back and forth, and it occurred to me that I could possibly do some good here (I am an avid reader of MacKenzie’s work and a big fan of this blog). If you want to understand the actual structural idiosyncrasies of the 2007 crisis, look up the series of papers that Yale’s Gary B. Gorton has been writing (Slapped in the face by the invisible hand, the panic of 2007, etc). Recommended by Ben Bernanke, no less. Gorton himself is a very interesting chap, a working class jazz head, who for a number of years was employed as a consultant by AIG to model the structured credit portfolios that they were selling CDS on.


  3. vimothy Says:

    On the interesting idea being floated above — but surely it would be hard to structure and price an MBS bond if the expected value of its tranches is in flux. Of course, this *was* the problem, but what would real time information, ceteris paribus, actually achieve?

    • marthapoon Says:

      vimotny, The idea imagines an attempt to make prices reflect all available information as per the theory of efficient markets. Its a thought experiment that extends a certain logic that is performed in financial markets. But you are certainly right to observe that how real time flows of information, price calculation and product design could be engineered into a working system, is another question…

  4. Chris Jefferis Says:

    Martha, I hadn’t thought of this idea as an extension of the logic of efficient markets. Its a good distillation.

    I’m having fun with this thought experiment.

    In practice I think it would take two different measures to put the thing into practice. One would be applying a VAR like calculation to the expenditure patterns of the mortgage holders underpinning the MBS to do continuing real time risk assessment. This means embedding VAR into instruments rather than as a regulatory apparatus of the institution.

    It would also be easier to implement this if all of the mortgage holders financial dealings (credit card, investments, bank account) were with a single institution say BofA because that would ensure that the MBS is underpinned by better information about expenditure.

    • marthapoon Says:

      I agree that the two suggests you make follow forward *logically* from the premise we’ve set out.

      I wouldn’t necessarily promote these positions given that the overall design would end up sustaining value in the interests of investors.

      Do any of these provisions protect the safety and fairness of the system for consumers…? It seems to me the system needs to be somehow equitable, not just logical, if it is to be stable.

  5. Chris Jefferis Says:

    I don’t think I’m promoting this idea (see my initial comments on privacy). I’m just interested in where it goes.

    Part of the issue for me is with which groups this logic of risk based pricing is applied to. Which groups are subject to a scheme for mortgage finance whose overall design is to sustain value in the interests of investors?

    I think those with poor credit ratings are more likely to have this logic applied to them because excessive monitoring of their finances is justified by their poverty. I’m interested in this power dynamic. Who ends up wearing the logic of efficient markets?

  6. […] September 21, 2009 Rob Wosnitzer conceived of the argument in this post and wrote it together with Martha Poon. It responds to the discussion of how real time information could contribute to MBS evaluation at the end of the previous post. […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: