Re:Nate Silver confuses cause and effect, ends up defending ...
by 001sky on Friday December 21, 2012 @03:31AM
(#4951995)
You seem to miss the point of Silver's conflict of interest. He is a Modeler, and has a financial gain in promoting certain views. Among these are (1) the notions that modelers are in general "experts"; and (2) that the intentions of people using the "product" of (1) are benign, or at any rate are not Opportunist. His PR agent would never let him say otherwise, it would be bad for business.
The financial crisis, at its heart was not an issue of "mistaken models". It was driven by (1) bad legislation from washington; and (2) unethical opportunism exploiting #1.
As a general rule, an Analyst at a Credit rating agency is a lowly position in wall street. Most important people (people acting of their own free will) disregard Rating Agency "analysis" out of hand. They do their own work. And people don't get promoted into Goldman Sachs, for example, from S&P et al very often. There is almosts a social stigma attached to the job, that would need to be cleansed with an MBA or some other "success" to pave the way. The people that cannot disregard rating agency work (due to law), of course are a built in market. And in general they are not on an even playing field with Wall Street with respect to their talent pool and access to information. So, bad analysis merely fills a vaccuum.
Using this as one example, think of the consequences. First, the model/analysis of the rating agency is a "product" looking to be sold (like a used car). There are no buyers for even the best analysis (wall st does its own research). So the "buyers" are not wall street but people who are forced to rely on credit ratings by law (eg, pension funds or some other public actors). But this is a captive market. The goal of the rating agency then becomes to game the system by extending the market. That is, creat new "protected classes" or create new "Asset types" that are salable to protected classes of investor. Structurally, now, this is the "game". The models, data-sets, and actual implementaions are really beside the point, provided there is a lack of transparency ("black box", proprietary, just complicated, etc). The models will be reverse engineered to "work", they only need to be "plasuble" (as in plausible deniability). Remember, the rating agencies are <<not legally experts>>. If they were, they could be sued, They are just expressing their 1st amendment rights <<to have an opinion>>.
None of this you seem to address in your commentary.
Nate Silver is just one in a long line of "experts" looking for ways to "sell their services". The first step in doing this is to never undermine the notion of "experts", or the idea that the system is driven by "expert knowledge". As the financial crisis has shown, those people who know what they are doing in the world are operating well beyond the scope and bounds of information that is available to people who publicly claim these hats. There is a reason hedge funds are famously secretive.
The financial crisis, at its heart was not an issue of "mistaken models". It was driven by (1) bad legislation from washington; and (2) unethical opportunism exploiting #1.
As a general rule, an Analyst at a Credit rating agency is a lowly position in wall street. Most important people (people acting of their own free will) disregard Rating Agency "analysis" out of hand. They do their own work. And people don't get promoted into Goldman Sachs, for example, from S&P et al very often. There is almosts a social stigma attached to the job, that would need to be cleansed with an MBA or some other "success" to pave the way. The people that cannot disregard rating agency work (due to law), of course are a built in market. And in general they are not on an even playing field with Wall Street with respect to their talent pool and access to information. So, bad analysis merely fills a vaccuum.
Using this as one example, think of the consequences. First, the model/analysis of the rating agency is a "product" looking to be sold (like a used car). There are no buyers for even the best analysis (wall st does its own research). So the "buyers" are not wall street but people who are forced to rely on credit ratings by law (eg, pension funds or some other public actors). But this is a captive market. The goal of the rating agency then becomes to game the system by extending the market. That is, creat new "protected classes" or create new "Asset types" that are salable to protected classes of investor. Structurally, now, this is the "game". The models, data-sets, and actual implementaions are really beside the point, provided there is a lack of transparency ("black box", proprietary, just complicated, etc). The models will be reverse engineered to "work", they only need to be "plasuble" (as in plausible deniability). Remember, the rating agencies are <<not legally experts>>. If they were, they could be sued, They are just expressing their 1st amendment rights <<to have an opinion>>.
None of this you seem to address in your commentary.
Nate Silver is just one in a long line of "experts" looking for ways to "sell their services". The first step in doing this is to never undermine the notion of "experts", or the idea that the system is driven by "expert knowledge". As the financial crisis has shown, those people who know what they are doing in the world are operating well beyond the scope and bounds of information that is available to people who publicly claim these hats. There is a reason hedge funds are famously secretive.