Disasters Class: In-Class Exercise
For class today, you listened to a podcast from NPR’s This American Life and Planet Money that sought to explain mortgage-backed securities and the housing crisis. You also read an article about the recent problems at Knight Capital.
Today in class you will think about these two events to try to come up with potential solutions to the problems presented. The catch is that the solutions cannot rely solely on the action of national governments, or on the actions of consumers (in other words, a “buyer beware” model). You should also include a brief explanation of why those two models of solving the problem wouldn’t be ideal.
Before you present that orally in class, however, I would like you to fire up your neurons by posting a brief paragraph (no more than 4 sentences) on how the podcast and the article are related. The goal will be to post that quickly enough so that your classmates can read it while they’re working on the rest of the assignment.
Note: if you were not present in class today, you should write out an answer to all of the above and post it in the comments in order to get credit for the 4th blog assignment.
With Anthony Cutuli and David Binnion
Perhaps the most interesting similarity between the mortgage-based securities and the Knight Capitol debacle is the fact that both relied on complex system that couldn’t be fully tested prior to their application in the real world. The biggest issue with both situations was that the systems being used, financial or computer, were not well understood by the people using them. Each worked well enough in theory, but every complex system is bound to have bugs. These bugs were never accounted for and no contingency was implemented.
In both cases, ignoring a problem for profit led to consumers being endangered. For the market collapse, they knowingly loaned money to consumers who could not pay back said loans, and with Knight Capital, no attempt to account for coding bugs was made.
The user of the code and the process needs to be better infromed; for example the process of how loans were distrubted banks were flawed since they didn’t want ask more information. In the Knight Capital case, code wanted to be viewed as a solution to every problem, when the programmers said there had to be “attentive humans” to help with code bugs.
The Knight Capital Group had computer errors causing flaws in stock prices which affected many companies. The housing bubble loans were considered profitable when computer simulations were ran but they were ran using the wrong data causing the loans to be continuously made and sold. A system must be in place to stop errors from becoming economic crises.
In both cases new advances in technology and finance lead to new, poorly understood problems. Since multiple parties are involved in the “final product” it becomes extremely difficult to identify the root cause. The advances bring financial profit and therefore certain flaws are overlooked along the way.
Both the article and the podcast addressed the issues of complex systems and how problems in these systems could cascade and grow into bigger problems. Where one part of the system failed and individuals tried to fix their problem, such as in the case of smaller banks trying to sell off the remainder of the NINA loans when other companies refused to buy them, the “fixes” only caused other problems to arise in other parts of the market. This is also seen in the debugging process in computer systems as mentioned in the New York Times article. If there had been some sensible human beings overseeing the whole system, rather than only being concerned about personal gain, then perhaps the issues caused by bad loans in the credit crisis and bug fixes leading to other bugs by Knight Capital programs.
Both situations rely on separate smaller systems to come together and create an operating and efficient larger system. Success in the smaller systems does not necessarily correlate to success when they come together. Thus identifying errors in the resulting larger system is subject to a thorough understanding of how the different parts interact and not just within their separate entities. In the Knight Capital situation the lack of knowledge surrounding software changes led to their financial loss. Likewise, the mortgage loan crisis is the result of insufficient knowledge regarding the consumers financial standing.