Tuesday, May 24, 2011

Lemons problem

Moral: Wherein we nod to an Ivy school (not Harvard), in the form of Sanjeev Arora, Boaz Barak, Markus Brunnermeier, Rong Ge, for offering insights about financial games that are remarkably accurate, in our opinion, and really bear on the issues (12/11/12 -- see: FAQ, at Princeton).


The title? Yes, if the buyers know that sellers only have 'lemons' (not, in any way, disparaging the wonderful fruit), then buyers will not buy (without heavy discounts) and the game ends. To whom will sellers sell?

Did we not see that with the recent affair where the bankers (and other financial types) essentially froze their gaming as they knew that they were dealing mainly with crooks (like themselves).

Liquidity left; the economy dumped. What happened?

Big Ben, and his ilk, put savers, and taxpayers, on the line to provide a huge pile of wealth so that these idiots would get back to their gaming. And, folks, we have not yet unwound from that. The chimera's rise is due, essentially, to our monies being offered to those who crapped on the economy (diaper changing, folks).


The 'lemon' concept is used in a very timely paper that covers some of the issues brought by computational support for finance. We have harped about that (many times).

(ACM Communications): Computational Complexity and Information Asymmetry in Financial Products. Abstract: Securitization of cash flows using financial derivatives transformed the financial industry over the last three decades. Derivatives have attracted criticism, but others say problems with derivatives would disappear with use of more accurate financial models, more vigilance by buyers and better governmental oversight.


The authors talk about financial derivatives, their use, and the troubles. The example derivatives are based upon mortgages, which is a timely subject. The use, supposedly, is because we can: modern environment, global scope, computers, mathematics, oodles of money, lots of suckers, and so on. The troubles?

Well, they are several, and we'll get technical about these later. For now, let's just itemize them with some comment.
  • The lemons problem - as mentioned above, this problem is why auto dealers offer a warranty. And, we might add, states enacted 'lemon' laws.
  • Complexity - ah, dear to our hearts, as undecidability lurks, everywhere. I like the example that they use for intractability -- which then leads to the decision problem, as how can you decide when you cannot compute? Oh, intuition? Yes, folks, as an aside, one job, in the future, will be trained intuitionistic overseers (ah, have we not tried that from the beginning? - no, the computer turned things around - think quasi-empiricism and Chaitin, et al). But, in particular, they show how it is much easier to check an answer than it is to actually find an answer.
  • CDOs - as a form of derivative (remember, Buffet said that these things were WMDs) for which there may be some value to we, the people - and, not via our bailing out the idiots. They talk tranches (trash, as we explained earlier).

Their conclusions? Well, for one, how do you determine that you're not getting crap, after the fat cat 'cherry picked' out the best? That is, the 'wedge' (difference between what the banker who wants to sell you junk thinks something is worth and what you can discern with your limited information) can be complicated to discover (actually, we deal with this type of thing as a general rule (such as, measuring progress in any of our endeavors), yet do so well when the cards are not stacked against us as we find with the current financial game).

In short, ex ante, even with the best of efforts, is not 20-20. We know that. One solution? Simple living folks driving our money system - it can be done (the military personnel who put their lives at stake are an example - not the best and brightest).

Then, transparency (no, dark pools, idiots -- by the way, this needs attention, too).

Even, ex post can be a problem. Yes. Who the hell has gone to jail out of the rogue table?

The paper is well worth the read. We'll go back through this whole thing, with some technical focus (adding to M&M and ergodic states).


06/11/2013 -- CDOs and tranching, once again.

12/22/2012 -- Fair and open actually used in a WSJ article.

12/13/2012 -- In the 12/11/2012 Remarks, the use of barbarian was in the context of either migratory (or invasive) movement of people from one locale to another with the result of the populace in the receiving bit of land undergoing an adjustment that could range from minor nuisance to major upheaval and death. That, then, motivates a look at why there might be migration, such as being forced. Turns out that the Wikipedia editors have done a good job of collecting the instances that we know of: Diasporas.

12/11/2012 --  Rick asks of the new barbarians from a historical perspective. ACM Communications, this month, interviews Sanjeev. The issue is locked, however this FAQ covers the topic very well.

01/01/2012 -- Recently ran across the work of Kazimierz Dabrowski. We need to pay more attention to his theory on development. Yes, CEOs (and other takers) as immature (seriously, so).

12/05/2011 -- It's interesting how idiotic the supposedly smart can be. The real issue: the failings of an idiot have a small influence; the failings of the 'real idiots' has wide impact (and, in so many ways). Somehow, we muddle through.

08/30/2011 -- Essentially, we have financial piracy.

07/12/2011 -- See Salem Commoners for a continuation of the theme. Also, changed 'Jaime' to 'Jamie' (oh yes).

05/31/2011 -- Lil Timmy. What a guy!

05/29/2011 -- Fair dealing, can that be brought back? Was it ever?

05/28/2011 -- We'll put avatars to more use than just being glorified (hyper-dimensional) icons.

05/27/2011 -- It's good to see others raise questions: why are the too big still doing crazy things? Why were there not prosecutions? ... It's disconcerting to hear that the feds (as in our elected officials, and their appointees) allowed (are allowing) the bank's sleight-of-hand in order to not 'rock the boat' or to keep the ease for the fat cat (miscreant aristocat).

05/25/2011 -- What they're talking: How do we control financial sleight-of-hand, which may even be unconscious, driven by humanness? Is the 'lemon' the norm in finance (and its gaming)? We have to learn how to 'engineer' truth, thanks to the growing prowess of computation in the hands of the idiots.

05/25/2011 -- The referenced article is under controlled access. However, here is an editorial review (appears in the magazine as a one-pager right before the article) that is available.

05/24/2011 -- How many times did we hear bankers say that they weren't going to lend? Despite all sorts of jawboning. Well, we could have nationalized (what does that mean?) the game more than we have so far with the hands-off approach (oh, they're adults, can self-govern - hah!!)? Their not lending is like the kid who takes his ball and won't play the game (so obvious, yet do we see any embarrassment, at all, of recognition on their part of their immaturity?).

Modified: 06/11/2013

No comments:

Post a Comment